Use of the Central Sensitization Inventory (CSI) as a treatment outcome measure for patients with chronic spinal pain disorder in a functional restoration program

Accepted Manuscript Title: Use of the central sensitization inventory (CSI) as a treatment outcome measure for chronic spinal pain disorder patients in a functional restoration program Author: Randy Neblett, Meredith M. Hartzell, Mark Williams, Kelley R. Bevers, Tom G. Mayer, Robert J. Gatchel PII: DOI: Reference:

S1529-9430(17)30270-X http://dx.doi.org/doi: 10.1016/j.spinee.2017.06.008 SPINEE 57348

To appear in:

The Spine Journal

Received date: Revised date: Accepted date:

18-10-2016 15-5-2017 7-6-2017

Please cite this article as: Randy Neblett, Meredith M. Hartzell, Mark Williams, Kelley R. Bevers, Tom G. Mayer, Robert J. Gatchel, Use of the central sensitization inventory (CSI) as a treatment outcome measure for chronic spinal pain disorder patients in a functional restoration program, The Spine Journal (2017), http://dx.doi.org/doi: 10.1016/j.spinee.2017.06.008. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

USE OF THE CENTRAL SENSITIZATION INVENTORY (CSI) AS A

2

TREATMENT OUTCOME MEASURE FOR CHRONIC SPINAL PAIN

3

DISORDER PATIENTS IN A FUNCTIONAL RESTORATION PROGRAM

4

Randy Neblett, M.A., L.P.C., BCB*

5

Meredith M. Hartzell, Ph.D. †

6

Mark Williams, Ph.D.*

7

Kelley R. Bevers, Ph.D. Candidate†

8

Tom G. Mayer, M.D.**

9

Robert J. Gatchel, Ph.D., ABPP†

10 11

*PRIDE Research Foundation, Dallas, TX

12 13

**Department of Orthopedic Surgery, University of Texas Southwestern Medical Center at Dallas, Dallas, TX

14 15

†Department of Psychology, University of Texas at Arlington, College of Science, 301 Life Science Building, Arlington, Texas 76019

16 17 18 19 20 21 22 23 24 25 26

Corresponding Author: Tom G. Mayer, M.D. 5701 Maple Ave. #100 Dallas, TX 75235 Phone:(214)351-6600 Fax: (214) 351-3026 E-mail:[email protected]

ABSTRACT

27

BACKGROUND CONTEXT: The Central Sensitization Inventory (CSI) is a valid and reliable

28

patient-reported instrument designed to identify patients whose presenting symptoms may be

29

related to Central Sensitization (CS). Part A of the CSI measures a full array of 25 somatic and

30

emotional symptoms associated with CS, and Part B asks if patients have previously been 1 Page 1 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

diagnosed with one or more specific Central Sensitivity Syndromes (CSSs) and related disorders.

2

The CSI has previously been validated in a group of chronic pain patients who were screened by

3

a trained psychiatrist for specific CSS diagnoses. It is currently unknown if the CSI can be a

4

useful treatment-outcome assessment tool for chronic spinal pain disorder (CSPD) patients who

5

are not screened for comorbid CSSs. It is known, however, that previous studies have identified

6

CS-related symptoms, and comorbid CSSs, in subsets of patients with CSPDs. Studies have also

7

shown that CS-related symptoms can be influenced by cognitive and psychosocial factors,

8

including abuse history in both childhood and adulthood, sleep disturbance, catastrophic and

9

fear-avoidant cognitions, and symptoms of depression and anxiety.

10

PURPOSE: To evaluate CSI scores, and their associations with other clinically-relevant

11

psychosocial variables, in a cohort of CSPD patients who entered and completed a functional

12

restoration program.

13

STUDY DESIGN/SETTING: A retrospective study of prospectively-collected data from a

14

cohort study of CSPD patients who completed the CSI at admission to, and discharge from, an

15

interdisciplinary function restoration program (FRP).

16

PATIENT SAMPLE: A cohort of 763 CSPD patients

17

OUTCOME MEASURES: Clinical interviews evaluated mood disorders and abuse history. A

18

series of self-reported measures evaluated comorbid psychosocial symptoms, including pain

19

intensity, pain-related anxiety, depressive symptoms, somatization symptoms, perceived

20

disability, and sleep disturbance, at FRP admission and discharge.

21

METHODS: Patients were grouped into five severity level groups, from Mild-to-Extreme, based

22

on total CSI scores, at FRP admission, and then again at discharge. The FRP included a

2 Page 2 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

quantitatively-directed and medically-supervised exercise process, as well as a multimodal

2

psychosocial disability management component.

3

RESULTS: The CSI severity groups were strongly associated with Major Depressive Disorder

4

and previous abuse history (p < .01), which are known risk factors for CS-related symptoms and

5

diagnoses. CSI scores were also strongly associated with patient-reported CSS diagnoses on CSI

6

Part B. The percentage of patients who reported a comorbid CSS diagnosis increased in each

7

higher CSI-severity group, from 11% in the Subclinical group, to 56% in the Extreme group. The

8

CSI severity groups were significantly related to other CS-related patient-reported symptoms,

9

including pain intensity, pain-related anxiety, depressive symptoms, somatization symptoms,

10

perceived disability, and sleep disturbance (ps < .001). CSI scores, along with all other

11

psychosocial measures, decreased at treatment discharge.

12

CONCLUSIONS: In the present study, admission CSI scores were highly associated with

13

previous CSS diagnoses, CS-related symptoms, and clinically relevant patient-reported

14

psychosocial variables. All psychosocial variables, as well as scores on the CSI, were

15

significantly improved at FRP discharge. The CSI may be have important clinical utility, as a

16

screener and as a treatment outcome measure, for CSPD patients participating in an

17

interdisciplinary FRP.

18 19

Keywords: central sensitization inventory, CSI, central sensitization, central sensitivity

20

syndrome, chronic spinal pain disorder, chronic pain, functional restoration program

21

3 Page 3 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

INTRODUCTION

2

Central Sensitization (CS) is a relatively new concept, which is gaining wide acceptance as a

3

functional dysregulation in the central nervous system, resulting in nociceptive hyperexcitability

4

and a lowered threshold for perception of sensory information, which amplifies pain and other

5

clinical symptoms [1]. Dysregulation in both ascending and descending CNS pathways, as a

6

result of potential genetic factors, physical traumatic injuries, and/or emotional distress, have

7

been proposed to explain the development and maintenance of CS [2-6]. Patients with CS often

8

display increased sensitivity to painful stimuli (i.e., hyperalgesia), pain in response to normally

9

non-painful stimuli (i.e., allodynia), expansion of the receptive field, and multifocal pain in

10

various body regions [7]. CS-related symptoms have also been found to be associated with abuse

11

history [8-11], psychiatric comorbidities (including Major Depressive Disorder and anxiety

12

disorders) [12, 13], and related symptoms (including sleep disturbance, fatigue, cognitive

13

deficits, opioid use, and fear-avoidant cognitions), that can result in inactivity, functional

14

limitations and decreased quality of life [14-21]. The term Central Sensitivity Syndrome (CSS)

15

describes a group of overlapping conditions that share a common pathophysiological mechanism

16

of CS [14, 22]. The CSS family of disorders includes fibromyalgia, chronic fatigue syndrome,

17

irritable bowel syndrome, migraine and tension headache, temporomandibular joint disorder, and

18

others [14]. CSS diagnoses are frequently associated with musculoskeletal complaints, whether

19

injury-related or non-injury-related in origin.

20

The Central Sensitization Inventory (CSI) was developed as a brief and easy-to-administer

21

patient-reported screener for patients who are at high risk of having CSSs [23]. Part A measures

22

a full array of 25 somatic and emotional symptoms associated with CS and CSSs, with a score

23

range of 0 to 100. Part B asks if patients have previously been diagnosed with specific CSSs and 4 Page 4 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

other CSS-related diagnoses. Although the original English version was only published in 2012

2

[23], the CSI is now available in Dutch [24, 25], Spanish [26], French [27], Guajarati [28],

3

Brazilian Portuguese [29], and Serbian [30], and is currently being translated into many other

4

European, Asian, and South American languages (personal communications). Thus, the clinical

5

interest in the CSI is quickly expanding, which speaks highly of its potential clinical utility.

6

Studies on multiple language versions of the CSI have shown good psychometric properties for

7

test-retest reliability (ICC ranging from 0.88 to 0.97) and internal consistency (Cronbach’s alpha

8

ranging from 0.88 to 0.91) [23, 25-31]. CSI total scores have been shown to discriminate

9

between chronic pain and control subjects [23, 25, 27, 29, 30], and between chronic pain patient

10

subgroups with, and without, objective CS-related diagnoses [9, 16, 27, 29, 30]. CSI scores have

11

also been found to be associated with CS-related symptoms [32-34]; objective biological markers

12

of CS [29, 35]; and patient-reported symptoms on other pain-related and validated questionnaires

13

[9, 36]. A cut-off score of 40 out of 100 on the CSI has demonstrated good sensitivity (81%) in

14

correctly identifying a group of CSS patients, and acceptable specificity (75%) in correctly

15

identifying a group of non-patient comparison subjects [16]. One recent study also found that

16

patients who scored above 40 on the CSI, before participating in knee arthroplasty, reported

17

more severe post-surgical pain intensity, required higher dosage of post-surgical analgesics, and

18

were at higher risk of persistent pain three months later [32]. This 40 point cut-off score has also

19

been recommended as one component of an algorithim to help classify chronic pain patients [37]

20

and low back pain patients [38] with CS, and to help differentiate them from patients with

21

neuropathic and nocioceptive pain. Most recently, five severity levels have been developed to

22

help aid in the clinicial interpretation of the CSI [36].

5 Page 5 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

It is still unknown whether the CSI can be a clinically useful patient-reported outcome (PRO)

2

measure for chronic spinal pain disorder (CSPD) patients, who have not been evaluated for a

3

CSS diagnosis. However, a number of previous studies have demonstrated that chronic spinal

4

pain is often related to CS [31, 39]. Investigations have found evidence for both local and

5

generalized hyperalgesia, and altered brain functioning, in response to various somatosensory

6

stimuli, including experimentally-induced pain, in subsets of CSPD subjects [40]. Spinal traumas

7

(e.g., motor vehicle collision, work injury, surgery) often trigger or worsen CS-related

8

symptoms, including chronic widespread pain [41-43]. One longitudinal study found that, as

9

regional spinal pain became more chronic, one was more likely to develop widespread pain [8].

10

Spinal traumas are also associated with the development of fibromyalgia, of which chronic

11

widespread pain is a primary symptom [5, 44, 45]. Two separate previous studies by our group,

12

in the same clinic as the present study, found that 32% of CSPD patients also met the American

13

College of Rheumatology diagnostic criteria [46] for chronic widespread pain [47], and 23% of

14

chronic musculoskeletal pain disorder patients met criteria for comorbid fibromyalgia [48]. A

15

separate study by another group found that 41% of patients entering a tertiary-level chronic pain

16

clinic met criteria for fibromyalgia [49]. Evidence suggests that for some CSPD patients, central

17

reorganization, leading to altered central nociceptive processing, can maintain pain in the

18

absence of ongoing peripheral nociception [40].

19

The purpose of the present study was multifocal. First, associations among CSI scores and other

20

patient reported outcome (PRO) measures, assessing psychosocial factors known to occur in

21

CSPD patients (pain severity, pain-related anxiety, depressive symptoms, perceived disability,

22

insomnia, and somatization symptoms) were evaluated at admission and discharge from a

23

Functional Restoration Program (FRP). In addition, associations among CSI scores and abuse 6 Page 6 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

history, current psychiatric diagnoses, and patient-reported previous CSS diagnoses from CSI

2

Part B were determined. Finally, admission-to-discharge treatment responsiveness of the CSI,

3

and the other PRO measures, was examined in this CSPD cohort.

4

MATERIALS AND METHODS

5

Subjects

6

A consecutive cohort of 1,151 chronic pain disorder patients, who had sustained musculoskeletal

7

injuries, were referred to, and consented to, treatment in an interdisciplinary FRP between

8

February 2010 and July 2013. Of the total sample of 1,151 musculoskeletal injury patients, 388

9

were eliminated because they only had an injury to a non-spinal body part, and did not develop a

10

CSPD, leaving 763 CSPD patients for analyses at treatment admission. Of that sample, 154

11

patients did not complete the FRP, leaving 609 patients for analyses of treatment responsiveness.

12

At FRP discharge, 71 patients failed to complete the CSI, leaving 538 patients for evaluation of

13

discharge variables. Figure 1 provides the CONSORT flow chart of the sample.

14

-----------------------------------------------

15

INSERT FIGURE 1 ABOUT HERE

16

-----------------------------------------------

17 18

The majority of patients admitted to the FRP (93%) had work-related injuries covered

19

through state or federal workers’ compensation claims, and the remainder of the patients were

20

covered through group-health policies. All patients entered the program with severe functional

21

limitations and either partial or total work disability. Approximately 76% were not working at

22

admission, and the other 24% were working in a limited capacity. Diagnoses included sprains,

23

strains, contusions, fractures, dislocations, nerve and soft tissue injuries. Approximately 36% 7 Page 7 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

(n=273) of the patients had spinal surgery prior to entering the FRP. Of the surgical patients,

2

67% (n=184) had fusions (including 5 patients with artificial disc replacement), and 63%

3

(n=171) had discectomies/laminectomies. Note that some of these patients had both types of

4

surgeries. The average time between the date of injury and treatment admission was 19.5 months

5

(see Table 1 for additional demographic information).

6

--------------------------------------------------------------

7

INSERT TABLE 1 ABOUT HERE

8

--------------------------------------------------------------

9

FRP participation criteria included: 1) the musculoskeletal trauma occurred at least four months

10

prior to entry into the program; 2) chronic disability remained following acute conservative care

11

and/or secondary care; 3) surgery was not a clear option or had not provided resolution; 4) severe

12

pain and functional limitations persisted; and 5) patients were able to communicate in English or

13

Spanish. All participants who enrolled in the FRP consented to the collection of information for

14

treatment management and research purposes. All data used in the present study were part of the

15

patients’ standard medical files, and therefore, the study was granted exemption status by the

16

local Institutional Review Board. All patients also signed a Health Insurance Portability and

17

Accountability Act (HIPAA) authorization prior to program participation. All information was

18

protected by this HIPAA rule, and confidentially was maintained in the Institute’s database as

19

part of ongoing quality-assurance procedures. No funding sources were utilized in the

20

preparation of this manuscript, and the authors had no conflicts of interest.

21

Psychosocial and Demographic Assessment

8 Page 8 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

Demographic information was collected in a structured interview format upon admission to the

2

FRP, including: gender, age, ethnicity, type of injury, number of injured body parts, length of

3

disability (i.e., time between the date of injury and the date of admission), total temporary

4

disability (i.e., time not working between the date of injury and the date of admission), and pre-

5

admission surgery to the injured spinal area. Patient-reported history of CSSs was collected from

6

the CSI Part B. Mental health status and abuse history (including physical, sexual, and

7

psychological) were assessed by a licensed mental health professional via a structured Mental

8

Health Evaluation. Post-injury Axis I diagnoses of Major Depressive Disorder (MDD) and

9

Generalized Anxiety Disorder (GAD) were determined from DSM-IV-TR diagnostic criteria

10

[50].

11

Psychosocial Patient-Reported Outcome (PRO) Measures

12

In addition to the CSI, a number of other patient-reported psychosocial and pain assessments

13

were administered, which are commonly used to assess CSPD patients. Each of the psychosocial

14

variables assessed by these PRO measures has also been found to be associated with CSS

15

diagnoses and CS-related symptoms [14-21]. The Pain Visual Analogue Scale (Pain VAS),

16

scored from 0 to 10 [51], assessed each patient’s pain level. The Oswestry Disability Index

17

(ODI), scored from 0 to 100 [52], and the Pain Disability Questionnaire (PDQ) [53, 54], scored

18

from 0 to 150, both assessed perceived disability with daily activities. While these questionnaires

19

are similar, the PDQ measures psychosocial aspects of perceived disability as well, and thus both

20

measures were included. When faced with chronic pain and disability, many patients exhibit

21

depressive symptoms, so depressive symptoms were assessed with the Beck Depression

22

Inventory (BDI), scored from 0 to 63 [55]. Many CSPD patients also suffer from insomnia,

23

which can occur independently of pain and depression [56], so insomnia was assessed with the 9 Page 9 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

Insomnia Severity Index (ISI), scored from 0 to 28 [57]. Because CSPD patients often report

2

catastrophic and fear-related cognitions, the Pain Anxiety Symptoms Scale (PASS), scored from

3

0 to 100, examined pain-related anxiety [58]. Lastly, because CSPD patients often exhibit high

4

levels of somatization, somatization symptoms were assessed with the Patient Health

5

Questionnaire (PHQ) – Somatization Module, scored from of 0 to 13 [59]. For all of these

6

measures, higher scores indicate more severe symptoms.

7

Statistical Methods

8

All data were analyzed using SPSS v. 18. One-way and repeated-measures analyses of variance

9

(ANOVAs) were used to compare groups for continuous variables. Post-hoc tests for differences

10

between groups were computed using a Bonferroni correction to adjust for Type I error.

11

Independent Chi-Square tests (χ2) were used for categorical variables, with adjusted residual

12

analyses conducted to determine the cells most associated with χ2 significance. Partial eta

13

squared effect sizes were also evaluated with the following ranges: small=.01; medium = .06; large

14

= 0.14. Partial eta squared was chosen as a measure of effect size because it is widely utilized,

15

easy to calculate using SPSS, and provides an estimate of the total proportion of variability

16

attributable to a factor (in this case, CSI severity categories) and is also independent of sample

17

size [60]. 90% confidence intervals were calculated around the partial eta squared effect sizes

18

using an SPSS file developed by Karl Wuensch, available on his statistical resources webpage

19

http://core.ecu.edu/psyc/wuenschk/Statistics.htm. The 90% confidence intervals were reported,

20

as opposed to 95% or 99%, given that partial eta squared cannot be a negative value, and

21

calculations of confidence intervals above 90% can include zero, directly conflicting with

22

interpreting statistical significance achieved in prior testing, as described in detail in the Steiger

23

(2004) review on effect size and confidence intervals [61]. Even though the ANOVA F-test is 10 Page 10 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

one-sided in hypothesizing efficacy of treatment on the psychosocial variables, we wanted to

2

report a two-sided interval; level was determined by 100 (1 – 2α)% = 90% confidence interval

3

for the effect sizes [61]. This method allows for interpretation of the standard ANOVA F-test

4

with a one-sided hypothesis, and the maintenance of a Type I error rate equivalent to alpha.

5

Further, the 90% confidence intervals produce a more accurate range of values concordant with

6

the methodology and rationale outlined in previous literature and, as such, we expect 90% of the

7

interval estimates to include the population parameters for each measurement [61, 62]. In order

8

to predict CSI severity categories, a hierarchical binary logistic regression analyses was

9

performed. All assumptions were met for this analysis, including sample size. A post-hoc power

10

analysis, conducted with G*Power 3.1 [63, 64], specified the need for a minimum of 122 patients

11

for Chi-square analysis (requiring 31 patients per group). These requirements were met. Missing

12

data were dealt with in a pairwise fashion, with patients not included in each analysis if they

13

were missing data for the target variable.

14

RESULTS

15

CSI scores were normally distributed among the 763 CSPD patient cohort at FRP admission,

16

with a range from 1 to 99, and a mean of 46.7 (SD = 15.4). The total cohort was classified into 5

17

CSI severity groups [36] at treatment admission: Subclinical (14%, n = 106); Mild (18%, n =

18

137); Moderate (25%, n = 194); Severe (23%, n = 178); and Extreme (19%, n = 148). The 609

19

program completers were also categorized into severity groups, based upon admission CSI

20

scores, for some of the analyses: Subclinical (15%, n = 92); Mild (18%, n = 112); Moderate

21

(26%, n = 160); Severe (22, n = 134); and Extreme (18%, n = 111). At treatment discharge, 538

22

patients completed the CSI, and were again re-classified into the five severity groups based upon

23

discharge CSI scores: Subclinical (38%, n = 204); Mild (22%, n = 118); Moderate (17%, n = 93); 11 Page 11 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

Severe (14%, n = 75); and Extreme (9%, n = 48). As can be seen, the distributions of CSI scores

2

were very similar between the total cohort and the treatment completers. The score distribution at

3

discharge was noticeably different, with a skew towards lower severity.

4 5

Demographic and Occupational Variables

6

Table 1 summarizes the demographic and occupational comparisons among the severity level

7

groups. Significant group differences were found for: gender; length of disability (number of

8

months between injury and treatment admission); total temporary disability (number of months

9

not working between the injury and treatment admission); the number of injured spinal regions;

10

and the percentage of patients who had pre-admission spinal surgeries. Post-hoc analyses found

11

that patients in the higher CSI severity groups had more injured body parts and longer pre-

12

treatment lengths of disability. The Subclinical severity group had significantly less pre-

13

treatment surgeries and were more likely to be male than the higher severity groups. The

14

Extreme severity group had significantly more pre-treatment surgeries and were more likely to

15

be female than the lower severity groups.

16

--------------------------------------------------------------

17

INSERT TABLE 1 ABOUT HERE

18

--------------------------------------------------------------

19

Table 2 presents the associations among the CSI severity groups and DSM-IV-TR psychiatric

20

diagnoses and abuse history. Significant differences were found for major depressive disorder

21

(MDD) and abuse history, but not for generalized anxiety disorder (GAD). Patients in the Severe

22

group were more likely, and patients in the Subclinical group were less likely, to be diagnosed

23

with MDD. Patients in the Extreme severity group were also more likely to report abuse history.

12 Page 12 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

--------------------------------------------------------------

2

INSERT TABLE 2 ABOUT HERE

3

--------------------------------------------------------------

4

Psychosocial Patient-Reported Outcome (PRO) Variables

5

Table 3 presents the admission and discharge CSI scores for the treatment completers, and the

6

associations among the CSI symptom-severity level groups and the other psychosocial PRO

7

measures. Scores on the CSI improved significantly (i.e., decreased) from admission (45.8; SD =

8

15.4) to discharge (35.6; SD = 28.4). Also, post-hoc analyses found significant differences in

9

CSI scores among all five CSI severity groups, at both admission and discharge. Differences

10

were also evident among the CSI severity groups for each of the other PRO measures. As the

11

CSI symptom severity levels increased, the scores on almost all the other psychosocial PRO

12

measures increased in a “stair-step” pattern, including pain intensity, perceived disability,

13

depressive symptoms, sleep disturbance, pain-related anxiety, and somatization-related

14

symptoms. Significant differences in these PRO scores were found among the majority of the

15

CSI severity groups, with all ps <.001. All the effect sizes ranged from medium to high,

16

indicating strong results.

17 18

--------------------------------------------------------------

19

INSERT TABLE 3 ABOUT HERE

20

--------------------------------------------------------------

21

Table 4 presents the associations among the CSI severity groups and the patient-reported CSSs

22

from CSI-Part B, including restless leg syndrome, chronic fatigue syndrome, fibromyalgia,

23

temporomandibular joint disorder, migraine or tension headaches, irritable bowel syndrome, and

13 Page 13 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

multiple chemical sensitivity. Of the total 763 patient cohort, 209 (27%) reported that they had

2

previously been diagnosed with one or more of these CSSs. The percentage of patients who

3

reported any previous CSS diagnoses increased in a “stair-step” fashion, from 11% in the

4

Subclinical CSI group, to 56% in the Extreme group. Patients in the Extreme CSI severity group

5

were significantly more likely to report previous CSS diagnoses, while those patients in the

6

Subclinical CSI severity group were less likely to report previous CSS diagnoses. When

7

evaluating specific CSSs, patients in the Extreme CSI severity group were significantly more

8

likely, and patients in the Subclinical severity group were significantly less likely, to report a

9

previous diagnosis of fibromyalgia, chronic fatigue syndrome, irritable bowel syndrome, restless

10

leg syndrome, and migraine/tension headaches.

11

--------------------------------------------------------------

12

INSERT TABLE 4 ABOUT HERE

13

--------------------------------------------------------------

14

Psychosocial Patient-Reported Outcome (PRO) Variables at Treatment Discharge

15

Table 5 shows the PRO scores at treatment Discharge for the 538 patients who completed both

16

the FRP and the CSI at treatment discharge, organized into 5 CSI severity groups. As with the

17

CSI severity groups at Admission, post-hoc analyses revealed significant differences in CSI

18

scores among all five discharge severity groups. Significant differences were also found among

19

the severity groups in the other PRO measures, with higher CSI severity generally associated

20

with higher patient-reported pain intensity, perceived disability, depressive symptoms, sleep

21

disturbance, pain-related anxiety, and somatization-related symptoms, with all ps <.001. All of

22

the effect sizes were in the large range.

23

14 Page 14 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1 2 3 4

-------------------------------------------------------------INSERT TABLE 5 ABOUT HERE -------------------------------------------------------------DISCUSSION

5

The CSI was originally designed as a screener to identify patients whose presenting symptoms

6

may be related to CS, and may be indicative of a CSS [23], so that appropriate assessment and

7

treatment approaches could be initiated. However, it has not been previously evaluated as a

8

treatment outcome measure. Previous studies have shown that CSPD patients, who are

9

characterized by chronic pain, functional disability, and significant psychosocial distress, are at-

10

risk for developing CS, CSSs, and related symptoms [5, 8, 31, 39, 41-45, 47]. In the present

11

investigation, a sample of CSPD patients completed a battery of psychosocial questionnaires,

12

including the CSI, at FRP Admission, and then again at Discharge. Patients were placed into one

13

of five CSI severity groups, based upon total CSI scores, for further analyses at both Admission

14

and Discharge. Responsiveness of CSI scores to the FRP, and their associations with other

15

clinically-relevant psychosocial variables, were assessed.

16

A number of demographic variables differed among the five CSI severity level groups at FRP

17

Admission. The Extreme CSI severity group had a higher percentage of females, which is a

18

known risk factor for CSSs [48, 65, 66]. The Extreme severity group had a higher rate of pre-

19

admission surgeries, and the Extreme and Severe groups had a greater number of injured body

20

parts. The Severe group also had a significantly longer length of disability (number of months

21

between injury and treatment admission). Thus, patients who scored higher on the CSI at

15 Page 15 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

Admission were associated with more prolonged pain and disability, and greater physical trauma,

2

which have all been found to relate to CS and CSS-related symptoms [5, 8, 67, 68].

3

All patients completed a mental health assessment at FRP Admission. When examining DSM-

4

IV-TR diagnoses, those in the Severe CSI severity group were more likely, and those in the

5

Subclinical group were less likely, to be diagnosed with MDD. Previous studies have found a

6

high prevalence of depression and anxiety in patients with CSSs [12, 13, 69]. Patients in the

7

Extreme CSI severity group were also more likely to report abuse history. Previous studies have

8

found that patients with CSSs, especially fibromyalgia, were more likely to have experienced

9

physical or sexual abuse [10, 11]. Additionally, among chronic low back or neck pain patients,

10

individuals who had a history of abuse in adulthood were 2.5 times more likely to develop

11

chronic widespread pain (an indicator of CS and concomitant CSS-related symptoms) [8]. When

12

they had a history of abuse in childhood, the likelihood of developing chronic widespread pain

13

was increased by 73%.

14

As stated previously, studies have shown that CSSs and CS-related symptoms are associated

15

with pain severity, depressive symptoms, insomnia, perceived disability, pain-related anxiety,

16

and somatization [14-16, 68]. These symptoms were assessed with a battery of psychosocial

17

PRO measures, which are commonly used in chronic pain research and treatment settings, at the

18

beginning and conclusion of treatment. At both Admission and Discharge, the CSI severity

19

groups were highly related to the scores on all of these measures. The Subclinical CSI severity

20

group reported the least amount, and the Extreme severity group reported the greatest amount, of

21

these symptoms. When categorized by both Admission CSI scores and by Discharge CSI scores,

22

the majority of the scores on these other PRO measures increased as CSI severity levels

23

increased, in a “stair-step” pattern. 16 Page 16 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

All PRO measures, including the CSI, improved between admission and discharge,

2

demonstrating that this patient cohort responded positively to the FRP. Although FRP

3

responsiveness of CSI scores was previous unknown, responsiveness of the other patient-

4

reported symptoms was expected, as many earlier studies in our clinic have found similar

5

decreases in self-reported psychosocial symptoms after completion of an FRP [70-72]. The FRP

6

in the current study used an interdisciplinary biopsychosocial treatment approach, which has

7

been recommended for treating both general non-malignant chronic pain, including CSPDs [73],

8

and CS-related symptoms of chronic pain [1, 74].

9

Responsiveness of the CSI was also demonstrated by a change in the score distributions at FRP

10

Admission and at Discharge. The CSI score distribution, by the five severity levels at Admission,

11

was relatively normal, with a slight skew toward the higher severity levels. Approximately 32%

12

of subjects scored within the Subclinical and Mild severity ranges, and 68% scored in the

13

Moderate to Extreme ranges. The score distribution at Discharge, though, was skewed to the

14

lower severity ranges, with approximately 60% of subjects scoring in the Subclinical and Mild

15

severity ranges, and 40% scoring in the Moderate to Extreme ranges.

16

In addition to evaluating treatment responsiveness of the CSI, CSI score differences were

17

analyzed among the five severity groups. Severity levels for the CSI have only been recently

18

proposed [36]. The results from the present study provide evidence for the validity of these

19

severity levels. It was found that CSI scores for each CSI severity group were significantly

20

different than all the other severity groups at both Admission and Discharge from the FRP. This

21

excellent CSI score discrimination among the severity groups suggests that these severity cutoffs

22

can be a useful guide for clinicians and researchers in the clinical interpretation of CSI scores for

23

similar clinical populations. 17 Page 17 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

Little was previously known about the prevalence of specific CSSs in CSPD populations. An

2

earlier study by our group, in the same clinic as the present study, found that 32% of post-injury

3

CSPD patients also met the American College of Rheumatology diagnostic criteria for chronic

4

widespread pain [47], a common CS-related symptom. Another study by our group found that

5

23% of chronic musculoskeletal pain patients in the same clinic met criteria for comorbid

6

fibromyalgia [48]. The present investigation has provided new information to the scientific

7

literature, revealing that 30% of CSPD patients reported a previous diagnosis of one or more

8

CSSs on the CSI Part B, including 12% who reported a previous fibromyalgia diagnosis.

9

Moreover, the number of patients reporting a previous CSS diagnosis increased in a “stair-step”

10

fashion from the lowest to highest CSI severity groups, so that 56% of patients in the Extreme

11

CSI severity group reported a previous CSS diagnosis, compared to only 11% of patients in the

12

Subclinical group. Patients in the Extreme CSI severity group were significantly more likely to

13

report a previous diagnosis of fibromyalgia, chronic fatigue syndrome, irritable bowel syndrome,

14

restless leg syndrome, and migraine/tension headaches. The Subclinical severity group was

15

significantly less likely to report these diagnoses. Although the CSPD patients in the present

16

study were not diagnosed with CSSs by a physician, another previous study by our group found

17

that patient-reported CSS diagnosis history from the CSI Part B, particularly those with irritable

18

bowel syndrome, fibromyalgia, and migraine/tension headaches, showed good agreement with a

19

physician’s diagnosis of these common CSSs [16]. It is highly likely that a subset of the CSPD

20

patients in the present study would have also met diagnostic criteria for one or more CSSs.

21

As in any large study, there were some limitations in the present investigation. First, the presence

22

of CS or specific CSS diagnoses, based upon established diagnostic criteria, were not available in

23

the CSPD sample. While patients had the opportunity to report previous CSS diagnosis history 18 Page 18 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

on the CSI Part B, these diagnoses were not confirmed with a more objective diagnostic

2

assessment. Other limitations include potential issues of generalizability, and the fact that the

3

power analysis was conducted retrospectively, which may have created additional bias. Also,

4

generalizability is always a potential issue in studies of this kind. The CSI was utilized in the

5

present investigation with a CSPD population, characterized by chronic pain, significant

6

functional disability, and associated psychosocial overlay (e.g., depression and anxiety). The CSI

7

has been previously utilized by our group in chronic pain patients with specific CSS diagnoses

8

who were referred for psychiatric evaluation and treatment. Previous studies by other groups

9

have utilized the CSI in patient populations with general chronic musculoskeletal pain [26],

10

shoulder pain [34], knee pain [32, 33], fibromyalgia[27, 29, 30], osteoarthritis [29], myofascial

11

pain syndrome [29], chronic tension-type headache [29], migraine headache [35], and chronic

12

low back pain [28]. Future studies will have to determine if the CSI is a useful measure in other

13

patient populations. Nevertheless, the many statistically-significant results, with large effect sizes,

14

make this present study an important new contribution to the scientific literature.

15 16

CONCLUSIONS

17

CS is becoming increasingly recognized as a significant factor in many chronic pain conditions,

18

including subsets of patients with CSPDs [31, 39]. Cognitive and psychosocial factors, including

19

patient-reported symptoms of depression and anxiety, diagnosed Major Depressive Disorder,

20

abuse history, sleep disturbance, pain intensity, perceived disability, catastrophic and fear-

21

avoidant cognitions, and somatization are all known to be associated with CS and CSSs [14-20,

22

68]. Although the CSPD patients in the present study were not screened for CSS diagnoses by a

23

qualified health-care provider, CSI scores at FRP admission were highly associated with patient19 Page 19 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

reported previous CSS diagnoses on CSI Part B, as well as a clinician-diagnosed Major

2

Depressive Disorder, and abuse history. Admission and Discharge CSI scores were highly

3

associated with patient-reported psychosocial variables, including pain intensity, pain-related

4

anxiety, depressive symptoms, perceived disability, sleep disturbance, and somatization. CSI

5

scores, along with all other patient-reported measures, were responsive to the FRP and decreased

6

at treatment Discharge.

7

The validity of the CSI has been demonstrated in previous studies, through its associations with

8

other pain-related self-report measures, CS-related clinical symptoms and biological markers,

9

and discrimination among diverse chronic pain subgroups and pain-free control subjects [23, 25,

10

26, 27, 29, 30, 32, 33, 35]. The results of the current study offer further evidence for the validity

11

of the CSI. In addition, this was the first study to demonstrate that patient-reported CS-related

12

symptoms on the CSI can respond positively to interdisciplinary treatment. Since subsets of

13

CSPD patients are at-risk for developing CS, the CSI may offer some clinical utility as a screener,

14

and as a treatment outcome measure, for this patient population. As noted above, the many

15

statistically significant results found, with associated large effect sizes, make this an important

16

contribution to the scientific literature.

17

20 Page 20 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1

REFERENCES

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

1. 2.

3. 4.

5.

6. 7.

8. 9. 10. 11. 12. 13. 14. 15. 16.

17. 18. 19.

Adams, L.M. and D.C. Turk, Psychosocial Factors and Central Sensitivity Syndromes. Current Rheumatology Reviews, 2015. 11(2): p. 96-108. Staud, R., The neurobiology of chronic musculoskeletal pain (including chronical regional pain), in Fibromyalgia and Other Central Pain Syndromes, D.J. Wallace and D.J. Clauw, Editors. 2005, Lippincott Williams and Willkins: Philadelphia, PA. p. 45-62. Ren, K. and R. Dubner, Descending modulation in persistent pain: an update. Pain, 2002. 100(1– 2): p. 1-6. van Wijk, G. and D.S. Veldhuijzen, Perspective on Diffuse Noxious Inhibitory Controls as a Model of Endogenous Pain Modulation in Clinical Pain Syndromes. The Journal of Pain, 2010. 11(5): p. 408-419. Buskila, D. and R. Mader, Trauma and work-related pain syndromes: Risk factors, clinical picture, insurance and law interventions. Best Practice & Research Clinical Rheumatology, 2011. 25(2): p. 199-207. Sampalli, T., et al., Proposed model of integrated care to improve health outcomes for individuals with multimorbidities. Patient Prefer Adherence, 2012. 6: p. 757-64. Goldenberg, D.L., D.J. Clauw, and M.A. Fitzcharles, New Concepts in Pain Research and Pain Management of the Rheumatic Diseases. Seminars in Arthritis and Rheumatism, 2011. 41(3): p. 319-334. Kindler, L.L., et al., Risk Factors Predicting the Development of Widespread Pain From Chronic Back or Neck Pain. The Journal of Pain, 2010. 11(12): p. 1320-1328. Neblett, R., et al., Ability of the central sensitization inventory to identify central sensitivity syndromes in an outpatient chronic pain sample. Clinical Journal of Pain, 2015. 31(4): p. 323-32. Häuser, W., et al., Emotional, physical, and sexual abuse in fibromyalgia syndrome: A systematic review with meta-analysis. Arthritis Care & Research, 2011. 63(6): p. 808-820. Wilson, D.R., Health Consequences of Childhood Sexual Abuse. Perspectives in Psychiatric Care, 2010. 46(1): p. 56-64. Arnold, L.M., et al., Comorbidity of fibromyalgia and psychiatric disorders. Journal of Clinical Psychiatry, 2006. 67(8): p. 1219-25. Henningsen, P., T. Zimmermann, and H. Sattel, Medically unexplained physical symptoms, anxiety, and depression: a meta-analytic review. Psychosomatic Medicine, 2003. 65: p. 528-533. Yunus, M.B., Central sensitivity syndromes: a unified concept for fibromyalgia and other similar maladies. J Indian Rheumatism Assoc, 2000. 8: p. 27-33. Yunus, M.B., Fibromyalgia and overlapping disorders: the unifying concept of central sensitivity syndromes. Seminars in Arthritis and Rheumatism, 2007. 36: p. 339-356. Neblett, R., et al., The Central Sensitization Inventory (CSI): Establishing Clinically Significant Values for Identifying Central Sensitivity Syndromes in an Outpatient Chronic Pain Sample. The Journal of Pain, 2013. 14(5): p. 438-445. Burgmer, M., et al., Cerebral activation and catastrophizing during pain anticipation in patients with fibromyalgia. Psychosom Med, 2011. 73(9): p. 751-9. Gracely, R.H., et al., Pain catastrophizing and neural responses to pain among persons with fibromyalgia. Brain, 2004. 127(4): p. 835-843. Sjörs, A., et al., An increased response to experimental muscle pain is related to psychological status in women with chronic non-traumatic neck-shoulder pain. BMC Musculoskeletal Disorders, 2011. 12(1): p. 1-12.

21 Page 21 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

20. 21.

22.

23. 24. 25. 26. 27. 28.

29.

30.

31. 32.

33. 34.

35. 36. 37. 38. 39.

Vase, L., et al., Cognitive-emotional sensitization contributes to wind-up-like pain in phantom limb pain patients. PAIN®, 2011. 152(1): p. 157-162. Herbert, M.S., et al., Pain Hypervigilance is Associated with Greater Clinical Pain Severity and Enhanced Experimental Pain Sensitivity Among Adults with Symptomatic Knee Osteoarthritis. Annals of Behavioral Medicine, 2014. 48(1): p. 50-60. Kindler, L.L., R.M. Bennett, and K.D. Jones, Central Sensitivity Syndromes: Mounting Pathophysiologic Evidence to Link Fibromyalgia with Other Common Chronic Pain Disorders. Pain Management Nursing, 2011. 12(1): p. 15-24. Mayer, T.G., et al., The Development and Psychometric Validation of the Central Sensitization Inventory. Pain Practice, 2012. 12: p. 276-285. van Wilgen, P., et al. Central Sensitization Inventory (CSI): Nederlandse vertaling. 2013 [cited 2016 August 20]; Available from: http://www.paininmotion.be/CSI-consensusvertaling.pdf. Kregel, J., et al., The Dutch Central Sensitization Inventory (CSI): Factor Analysis, Discriminative Power, and Test-Retest Reliability. Clinical Journal of Pain, 2016. 32(7): p. 624-30. Cuesta-Vargas, A.I., et al., Cross-cultural adaptation and validity of the Spanish central sensitization inventory. SpringerPlus, 2016. 5(1): p. 1837. Pitance, L., et al., Cross cultural adaptation, reliability and validity of the French version of the central sensitization inventory. Manual Therapy, 2016. 25: p. e83-e84. Bid Dibyendunarayan, D., et al., Content Validity and Test-Retest Reliability of the Gujarati Version of the Central Sensitization Inventory. National Journal of Integrated Research in Medicine, 2016. 7: p. 18-24. Caumo, W., et al., Cross-Cultural Adaptation and Validation of the Central Sensitization Inventory for a Brazilian Population and its relationship with the function of descending pain modulatory system and with the serum BDNF. Journal of Pain Research, submitted for publication. Knezevic, A., et al., Cross cultural adaptation and psychometric validation of the Serbian version of the central sensitization inventory (CSI). Pain Practice: The Official Journal of World Institute of Pain 2017, Submitted for publication. Bid, D.D., N.C. Soni, and P.V. Rathod, Central Sensitization In Chronic Low Back Pain: A Narrative Review. NJIRM, 2016. 7(3): p. 114-123. Kim, S.H., et al., Influence of Centrally Mediated Symptoms on Postoperative Pain in Osteoarthritis Patients Undergoing Total Knee Arthroplasty: A Prospective Observational Evaluation. Pain Practice, 2015. 15(6): p. E46-E53. Lluch Girbés, E., et al., Expanded Distribution of Pain as a Sign of Central Sensitization in Individuals With Symptomatic Knee Osteoarthritis. Physical Therapy, 2016. 96(8): p. 1196-1207. Coronado, R., et al., (180) Pain sensitivity questionnaire differentiates individuals with shoulder pain who demonstrate widespread sensitivity with quantitative sensory testing. The Journal of Pain. 15(4): p. S21. Aguila, M.-E.R., et al., The Association Between Clinical Characteristics of Migraine and Brain GABA Levels: An Exploratory Study. The Journal of Pain, 2016. 17(10): p. 1058-1067. Neblett, R., et al., Establishing clinically relevant severity levels for the central sensitization inventory. Pain Practice, 2017. 17(2): p. 166-175. Nijs, J., et al., Applying modern pain neuroscience in clinical practice: criteria for the classification of central sensitization pain. Pain Physician, 2014. 17(5): p. 447-57. Nijs, J., et al., Low back pain: guidelines for the clinical classification of predominant neuropathic, nociceptive, or central sensitization pain. Pain Physician, 2015. 18(3): p. E-333-46. Sanzarello, I., et al., Central sensitization in chronic low back pain: A narrative review. Journal of Back Msuculoskeletal Rehabilitation, 2016. preprint: p. 1-9.

22 Page 22 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

40. 41. 42.

43.

44. 45. 46.

47.

48.

49. 50. 51. 52. 53.

54.

55. 56. 57. 58. 59.

Roussel, N.A., et al., Central sensitization and altered central pain processing in chronic low back pain: fact or myth? Clinical Journal of Pain, 2013. 29(7): p. 625-38. McBeth, J., et al., The role of workplace low-level mechanical trauma, posture and environment in the onset of chronic widespread pain. Rheumatology, 2003. 42(12): p. 1486-1494. Wenzel, H., A. Mykletun, and T. Nilsen, Symptom profile of persons self-reporting whiplash: a Norwegian population-based study (HUNT 2). European Spine Journal, 2009. 18(9): p. 13631370. Myrtveit, S.M., et al., Somatic symptoms beyond those generally associated with a whiplash injury are increased in self-reported chronic whiplash. A population-based cross sectional study: the Hordaland Health Study (HUSK). BMC Psychiatry, 2012. 12(1): p. 1-12. Waylonis, G.W. and R.H. Perkins, Post-traumatic fibromyalgia. A long-term follow-up. American Journal of Medical Rehabilitation, 1994. 73(6): p. 403-12. Buskila, D., et al., Increased rates of fibromyalgia following cervical spine injury. A controlled study of 161 cases of traumatic injury. Arthritis and Rheumatism, 1997. 40(3): p. 446-452. Wolfe, F., et al., The American College of Rheumatology 1990 Criteria for the Classification of Fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis and Rheumatism, 1990. 33(2): p. 160-72. Mayer, T.G., et al., Chronic Widespread Pain in Patients with Occupational Spinal Disorders: Prevalence, Psychiatric Comorbidity, and Association with Outcomes. Spine, 2008. 33: p. 18891897. Howard, K.J., et al., Fibromyalgia syndrome in chronic disabling occupational musculoskeletal disorders: Prevalence, risk factors and post-treatment outcomes. Journal of Occupational Environmental Medicine, 2010. 52(12): p. 1186-91. Brill, S., et al., Prevalence of fibromyalgia syndrome in patients referred to a tertiary pain clinic. Journal of Investigative Medicine, 2012. 60(4): p. 685-8. American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, 4th ed, Text Revision.2000: American Psychiatric Association. Capra, P., M. T.G., and R.J. Gatchel, Adding psychological scales to your back pain assessment. The Journal of Musculoskeletal Medicine, 1985. 2: p. 41-52. Fairbank, J.C., et al., The Oswestry low back pain disability questionnaire. Physiotherapy, 1980. 66: p. 271-273. Gatchel, R.J., T. Mayer, and R. Theodore, The Pain Disability Questionnaire: Relationship to oneyear functional and psychosocial rehabilitation outcomes. Journal of Occupational Rehabilitation, 2006. 16: p. 75-94. Anagnostis, C., R.J. Gatchel, and T.G. Mayer, The Pain Disability Questionnaire: A new psychometrically sound measure for chronic musculoskeletal disorders. Spine, 2004. 29: p. 22902302. Beck, A.T., et al., An inventory for measuring depression. Archives of General Psychiatry., 1961. 4: p. 561-571. Asih, S., et al., Insomnia in a chronic musculoskeletal pain with disability population is independent of pain and depression. The Spine Journal, 2014. 14(9): p. 2000-2007. Morin, C.M., Insomnia: Psychological Assessment and Management1993, New York, NY: Guilford Press. McCracken, L.M. and L. Dhingra, A short version of the Pain Anxiety Symptoms Scale (PASS-20): Preliminary development and validity. Pain Research & Management, 2002. 7(1): p. 45-50. Spitzer, R.L., K. Kroenke, and J.B.W. Williams, Validation and utility of a self-report version of PRIME-MD: The PHQ primary care evaluation of mental disorders. JAMA, 1999. 282(18): p. 17371744.

23 Page 23 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

60. 61. 62.

63. 64. 65. 66.

67. 68.

69. 70.

71.

72.

73.

74.

Levine, T. R, & Hullett, C. R. Eta Squared, Partial Eta Squared, and Misreporting of Effect Size in Communication Research. Human Communication Research, 2002. 28(4): p. 612-625. Steiger, J. H. Beyond the F Test: Effect Size Confidence Intervals and Tests of Close Fit in the Analysis of Variance and Contrast Analysis. Psychological Methods, 2004. 9(2): p. 164-182. Smithson, M. Correct confidence intervals for various regression effect sizes and parameters: The importance of noncentral distributions in computing intervals. Educational and Psychological Measurement, 2001. 61(4): p. 605-632. Faul, F., et al., G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavioral Research Methods, 2007. 39(2): p. 175-91. Faul, F., et al., Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behavioral Research Methods, 2009. 41(4): p. 1149-60. Shyti, R., B. de Vries, and A. van den Maagdenberg, Migraine Genes and the Relation to Gender. Headache: The Journal of Head and Face Pain, 2011. 51(6): p. 880-890. Voß, U., A. Lewerenz, and K. Nieber, Treatment of Irritable Bowel Syndrome: Sex and Gender Specific Aspects, in Sex and Gender Differences in Pharmacology, V. Regitz-Zagrosek, Editor 2012, Springer Berlin Heidelberg: Berlin, Heidelberg. p. 473-497. Diatchenko, L., et al., Idiopathic pain disorders – Pathways of vulnerability. Pain, 2006. 123(3): p. 226-230. Smart, K.M., et al., Self-reported pain severity, quality of life, disability, anxiety and depression in patients classified with ‘nociceptive’, ‘peripheral neuropathic’ and ‘central sensitisation’ pain. The discriminant validity of mechanisms-based classifications of low back (±leg) pain. Manual Therapy, 2012. 17(2): p. 119-125. Fischer, S., et al., Prevalence, Overlap, and Predictors of Functional Somatic Syndromes in a Student Sample. International Journal of Behavioral Medicine, 2013. 20(2): p. 184-193. Hartzell, M.M., et al., Do comorbid fubromyalgia diagnoses change after a functional restoration program in patients with chronic disabling occupational musculoskeletal disorders? Spine, 2014. 39(17): p. 1393-400. Hartzell, M.M., et al., Somatization as a predictor of outcomes following functional restoration of chronic disabling occupational musculoskeletal pain disorder patients. Journal of Applied Biobehavioral Research, 2013. 18(2): p. 59-81. Asih, S., et al., Does Classification of Chronic Musculoskeletal Disorder Patients Into Psychosocial Subgroups Predict Differential Treatment Responsiveness and 1-Year Outcomes After a Functional Restoration Program? The Clinical Journal of Pain, 2015. 31(12): p. 1036-1045. Gatchel, R.J. and A. Okifuji, Evidence-Based Scientific Data Documenting the Treatment and CostEffectiveness of Comprehensive Pain Programs for Chronic Nonmalignant Pain. The Journal of Pain, 2006. 7(11): p. 779-793. Nijs, J., et al., Treatment of central sensitization in patients with ‘unexplained’ chronic pain: an update. Expert Opinion on Pharmacotherapy, 2014. 15(12): p. 1671-1683.

39 40

24 Page 24 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients Figure 1. Consolidated Standards of Reporting Trials (CONSORT) Flow Chart of the Present Sample

25 Page 25 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients Table 1. Demographic and Occupational Comparisons by CSI Severity Groups at Treatment Admission (N = 763) Demographic/ Occupational Total Subclinicala 0- Mildb Moderatec Severed Variables N = 763 29 30-39 40-49 50-59 N = 106 N= 137 N= 194 N= 178

Extremee 60+ N= 148

Significance p value

Gender n (% Male)

493 (65%)

79 (75%) 2

92 (67%)

134 (69%)

111 (62%)

77 (52%) 1

.00

Age mean year (SD)

46.8 (10.6)

44.8 (11.6)

46.5 (10.9)

47.5 (10.8)

46.9 (9.9)

47.4 (10.2)

.27

** Length of Disability, mean number of months (SD)

19.5 (30.8)

15.4 (19.9)d

15.0 (23.6)d

19.3 (33.1)

26.1 40.2)a,b

25.8 (35.1)

.00

*** Total Temporary Disability, mean number of months (SD)

12.0 (20.1)

8.3 (8.7)

10.7 (14.7)

14.4 (29.3)

13.0 (19.2)

15.2 (21.9)

.03

Mean number of injured spinal regions (SD)

2.0 (1.6)

1.6 (1.3)d,e

2.0 (1.6)e

2.0 (1.6)e

2.2 (1.6)a,e

2.8 (1.9)a-d

.00

Number (n) and percentage (% yes) of subjects with surgeries

273 (36%)

26 (25%) 1

44 (32%)

68 (35%)

71 (40%)

64 (43%) 2

.02

Total number of surgical procedures • Discectomy/Laminectomy • Fusion/ Artificial Disc

355 171 184

35 1 21 14 1

50 22 28

86 40 46

94 42 52

90 46 44

.03 .52 .01

****Pre-admission Spinal Surgeries

1 – Significantly less likely 2 – Significantly more likely *Superscript letters indicate which groups significantly differed from each other in continuous comparisons ** Length of Disability indicates the total number of months between injury and admission, including partial disability from work (working with modification due to disability), and total disability from work (not working at all due to disability). *** Total Temporary Disability indicates the total number of months not working (e.g., total disability from work) between injury and admission. **** Note that some subjects had more than one type of surgery

26 Page 26 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients Table 2. DSM-IV Psychiatric Disorders and Abuse History Among the CSI Severity Groups at Admission (N = 763) Psychiatric Disorders, Subclinical0Mild Moderate Severe Extreme Abuse, and Previous CSSs 29 30-39 40-49 50-59 60+ N = 106 N= 137 N= 194 N= 178 N= 148

Significance p value

Axis I Disorder Major Depressive Disorder (n, % yes) Generalized Anxiety Disorder (n, % yes) Abuse History (n, % yes)

29 (27%)1

50 (37%)

73 (38%)

94 (53%)2

78 (53%)

.00

6 (6%)

17 (12%)

27 (14%)

25 (14%)

19 (13%)

.25

6 (6%)

10 (7%)

15 (8%)

22 (12%)

24 (16%)2

.02

1 = less likely 2 = more likely

27 Page 27 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients Table 3. Change in Psychosocial Patient-Reported Variables by CSI Severity Groups at Treatment Admission, Treatment Completers Only (N = 609) Psychosocial Whole Subclinicala Mildb Moderatec Severed Extremee Significance Effect Confidence Measures* Cohort 0-29 30-39 40-49 50-59 60+ p value Size Interval N = 92 N= 112 N= 160 N= 134 N= 111 ** Central Sensitization Inventory Mean (SD) Admission Discharge Change

45.8 (15.4) 35.6 (17.2) 10.2

22.1 (6.0)b-e 18.3 (9.9)b-e 3.8

35.3 (2.7)a,c-e 28.5 (13.6)a,ce

44.2 (2.9)a,b,d,e 33.9 (13.0)a,b,d,e 10.3

6.8 Pain VAS Mean (SD) Admission Discharge Change

Oswestry Disability Index (ODI) Mean (SD) Admission Discharge Change

Pain Disability Questionnaire(PDQ) Mean (SD) Admission Discharge Change

54.5 (2.8)a-c,e 42.0 (13.6)ac,e

68.6 (7.2)a-d 52.7 (15.9)a-d 15.9

.00 .00 .00

.92 .40 .08

[.91, .92] [.34, .44] [.05, .12]

12.5

7.4 (3.8) 5.0 (3.3) 2.4

6.3 (2.2)c,e 3.7 (3.3)c-e 2.5

7.1 (1.9) 4.7 (3.5) 2.4

7.8 (6.4)a 5.0 (2.5)a 2.7

7.5 (1.7) 5.2 (3.7)a 2.2

8.1 (2.1)a 5.9 (3.2)a 2.1

.01 .00 .87

.02 .04 .002

[.01, .04] [.01, .06] [.00, .00]

42.5 (18.7) 28.2 (19.9) 14.3

30.6 (18.7)d-e 16.0 (15.2)b-e 14.6

38.4 (17.1)a,d,e 23.7 (17.2)a,d,e 14.7

40.3 (16.6)a,e 28.8 (17.3)a,e 11.5

44.9 (19.4)a,b,e 32.4 (20.5)a,b 12.5

53.0 (17.0)a-d 37.1 (22.7)a-c 15.9

.00 .00 .30

.13 .11 .01

[.09, .17] [.07, .15] [.00, .02]

101.6 (23.8) 68.4 (36.5) 33.2

79.4 (26.9)b-e 45.2 (29.7)b-e 34.2

94.7 (21.9)a,c-

102.3 (19.5)a,b,e 70.1 (31.8)a,e 32.2

105.5 (20.7)a,b,e 75.1 (35.9)a,b 30.4

117.0 (19.5)a-d 86.1 (39.7)a-c 30.9

.00 .00 .67

.22 .13 .004

[.17, .26] [.08, .16] [.00, .01]

e

59.2 (33.7)a,d,e 35.5

28 Page 28 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients Psychosocial Measures*

Beck Depression Inventory (BDI) Mean (SD) Admission Discharge Change

Insomnia Severity Index (ISI) Mean (SD) Admission Discharge Change

Pain Anxiety Symptom Scale (PASS) Mean (SD) Admission Discharge Change

PHQ-Somatization Mean (SD) Admission Discharge Change

Whole Cohort

Subclinicala 0-29 N = 92

Mildb 30-39 N= 112

Moderatec 40-49 N= 160

Severed 50-59 N= 134

Extremee 60+ N= 111

Significance p value

Effect Size **

Confidence Interval

18.6 (10.4) 11.8 (10.4) 6.8

9.6 (6.9)b-e 4.9 (6.2)b-e 4.7

14.6 (7.5)a,d,e 9.3 (9.4)a,d,e 5.3

16.9 (8.0)a,d,e 11.1 (8.4)a,e 5.8

21.5 (9.1)a-c,e 13.2 (9.8)a,b,e 8.3

28.4 (10.2)a-d 19.1 (12.5)a-d 9.3

.00 .00 .00

.33 .17 .03

[.28, .37] [.12, .21] [.01, .05]

15.2 (10.9) 10.9 (9.7) 4.3

8.6 (9.5)b-e 5.8 (7.8)c-e 2.8

14.5 (11.8)a,e 9.6 (9.3)e 4.9

15.2 (9.6)a 11.8 (8.9)a 3.4

16.9 (10.6)a 11.8 (10.2)a 5.1

18.5 (10.6)a,b 14.0 (10.4)a,b 4.5

.00 .00 .25

.08 .07 .01

[.04, .11] [.03, .10] [.00, .02]

39.1 (27.3) 14.7 (25.2) 24.4

28.9 (21.8)e 6.5 (17.2)e 22.4

30.9 (22.2)e 8.4 (19.3)e 22.5

40.1 (26.6) 14.0 (25.6) 26.1

42.9 (25.1) 17.1 (23.0) 25.8

56.2 (31.1)a,b 26.3 (33.3)a,b 29.9

.00 .01 .84

.13 .08 .01

[.05, .19] [.01, .13] [.01, .02]

9.1 (4.5) 6.9 (5.6) 2.2

5.3 (2.9)b-e 3.9 (3.7)c-e 1.4

7.2 (3.5)a,d,e 5.3 (4.5)d,e 1.9

8.3 (3.6)a,d,e 6.9 (5.6)a,e 1.4

10.4 (4.1)a,b,d,e 8.2 (5.7)a,b 2.2

13.6 (3.8)a-d 9.4 (6.4)a-c 4.2

.00 .00 .00

.35 .11 .03

[.30, .39] [.07, .14] [.01, .05]

*Superscript letters indicate a significant difference between groups on the Bonferroni-adjusted post-hoc analyses * * Partial eta squared effect size (small=.01; medium = .06; large = 0.14)

29 Page 29 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients Table 4. Patient-Reported Previous CSSs on CSI Part B at Admission (N = 763) Patient-reported previous CSSs

Mild 30-39 N= 137 29 (22%)

Moderate 40-49 N= 194 50 (27%)

Severe 50-59 N= 178 52 (33%)

Extreme 60+ N= 148 66 (56%)2

Significance p value

Any previous CSS

Subclinical029 N = 106 12 (11%)1

Fibromyalgia

1 (1%)1

5 (4%)1

12 (7%)

22 (13%)

33 (24%)2

.00

Chronic Fatigue Syndrome

1 (1%)1

2 (2%)1

7 (4%)

18 (11%)

23 (16%)2

.00

Migraine/Tension Headaches

9 (9%)1

20 (16%)1

41 (22%)

48 (29%)

64 (45%)2

.00

Temporomandibular Joint Disorder

3 (3%)

5 (4%)

12 (7%)

14 (9%)

15 (11%)

.09

Restless Leg Syndrome

1 (1%)1

6 (5%)

18 (10%)

17 (10%)

25 (18%)2

.00

Irritable Bowel Syndrome

0 (0%)1

5 (4%)1

13 (7%)

19 (12%)

36 (26%)2

.00

Multiple Chemical Sensitivities

0 (0%)

0 (0%)

4 (2%)

5 (3%)

6 (4%)

.06

.00

1 = less likely 2 = more likely

30 Page 30 of 31

The Central Sensitization Inventory (CSI) For Chronic Spinal Pain Disorder Patients Table 5. Discharge Psychosocial Patient-Reported Variables by CSI Severity Groups at Treatment Discharge, Treatment Completers Only (N = 538) Psychosocial Measures* Total Subclinicala Mildb Moderatec Severed Extremee Significance Effect Confidence Cohort 0-29 30-39 40-49 50-59 60+ p value Size * Interval N = 204 N= 118 N= 93 N= 75 N= 48 * Central Sensitization Inventory Mean (SD)

35.6 (17.2)

18.2 (7.1)b-e

34.4 (2.9)a,

44.4 (2.9)a, b,

53.7 (2.9)a-c,

c-e

d, e

e

67.8 (8.1)a-d

.00

.90

[.89, .91]

Pain VAS Mean (SD)

5.4 (2.3)

4.4 (2.4)b-e

5.2 (2.2)a,c

6.2 (1.8)a,b

6.2 (1.9)a,b

6.9 (1.7)a,b

.00

.15

[.10, .19]

Oswestry Disability Index (ODI) Mean (SD)

29.8 (18.7)

17.9 (14.5)b-e

29.3 (13.6)a, c-e

37.1 (16.4)a,b,e

43.3 (16.4)a,b

47.7 (17.9)a,b,c

.00

.34

[.28, .38]

Pain Disability Questionnaire(PDQ) Mean (SD)

71.6 (33.5)

48.7 (28.8)b-e

71.9 (24.4)a, c-e

86.0 (27.5)a,b,e

95.1 (24.6)a,b

104.1 (26.4)a,b,c

.00

.36

[.31, .41]

Beck Depression Inventory (BDI) Mean (SD)

12.3 (9.9)

5.2 (5.1)b-e

11.9 (6.1)a,

15.6 (7.8)a, b,

19.5 (9.0)a-c,

26.2 (12.0)a-d

.00

.47

[.42, .51]

c-e

d, e

e

Insomnia Severity Index (ISI) Mean (SD)

11.9 (8.9)

7.3 (7.4)b-e

12.4 (8.5)a,

12.6 (8.2)a, d,

17.7 (6.1)a-c

19.2 (9.3)a-c

.00

.22

[.16, .26]

d, e

e

Pain Anxiety Symptom Scale (PASS) Mean (SD)

16.3 (25.5)

6.5 (17.6)a,

9.5 (23.1)a,

15.0 (24.5)d,

38.3 (30.8)a-c

46.4 (35.7)a-c

.00

.21

[.11, .28]

d, e

d, e

e

PHQ-Somatization Mean (SD)

7.5 (4.8)

4.4 (2.9)b-e

7.6 (3.4)a, d,

8.6 (3.2)a, d, e

11.1 (6.0)a-c

12.8 (5.0)a-c

.00

.35

[.30, .40]

e

*Superscript letters indicate a significant difference between groups on the Bonferroni-adjusted post-hoc analyses * * Partial eta squared effect size (small = .01; medium = .06; large = 0.14)

31 Page 31 of 31