Primary Image

Rehabilitation Measures Database

Pain Catastrophizing Scale

Last Updated

Purpose

The Pain Catastrophizing Scale (PCS) is a 13-item self-report measure designed to assess catastrophic thinking related to pain among adults with or without chronic pain.

Link to Instrument

Instrument Details

Acronym PCS

Area of Assessment

Cognition
Negative Affect
Pain
Stress & Coping

Assessment Type

Patient Reported Outcomes

Administration Mode

Paper & Pencil

Cost

Free

Actual Cost

$0.00

Key Descriptions

  • 13 items rated on 5-point Likert scales, from (0) not at all to (4) all the time
  • Higher score indicates higher level of catastrophizing
  • Item scores are summed into a total score (PCS-T) and three subscale scores:
    - Rumination (PCS-R): Items 8, 9, 10, and 11
    - Magnification (PCS-M): Items 6, 7, 13
    - Helplessness (PCS-H): Items 1, 2, 3, 4, 5, and 12
  • Minimum total score = 0, maximum total score = 52
  • Subscale score ranges: PCS-R: 0–16; PCS-M: 0–12; PCS-H: 0–24
  • A total score above 30 indicates clinically relevant level of catastrophizing
  • Individuals are instructed to reflect on past painful experiences and to indicate the degree to which they experienced each of the 13 thoughts or feelings when experiencing pain
  • Requires a reading level of approximately Grade 6

Number of Items

13

Equipment Required

  • Printed or electronic questionnaire

Time to Administer

5 minutes

Required Training

No Training

Age Ranges

Adult

18 - 64

years

Instrument Reviewers

Updated in 2020:

Spencer Brodsky, The George Washington University Medical School

 

Written in 2019:

Yunzhen Huang, MS, University of Wisconsin-Madison

Jing Cao, MS, MEd, University of Wisconsin-Madison

Susan Miller Smedema, PhD, CRC, LPC, University of Wisconsin-Madison

ICF Domain

Participation

Measurement Domain

Cognition
Emotion

Considerations

  • The PCS requires a reading level of approximately Grade 6.
  • Chronic pain clinical sample scores higher than the nonclinical sample, especially on the rumination subscale.
  • Females score higher than males in both chronic pain clinical sample and nonclinical community sample.
  • The percentile scores in the PCS user manual were generated from a sample of Canadian injured workers, among which 75% had soft tissue back injury. Therefore, the sample may not be representative of general chronic pain samples. In addition, percentile scores should be interpreted with caution when applied to nonclinical samples or individuals with acute pain.
  • The PCS has various adaptations that have been validated in the literature. For adults, there is a significant-other proxy version. For children, there is a child version and a parent proxy version. The PCS has also been translated and validated in 24 languages.
  • A Systematic Review and Meta-Analysis performed reports excellent internal reliability and test-retest reliability for the total PCS, but not for the subscales (Wheeler, 2019).

Chronic Pain

back to Populations

Standard Error of Measurement (SEM)

Chronic Pain: (calculated from Osman et al., 2000; n=60; mean age = 32.2 (9.8) years)

  • SEM for entire group (n = 60): 2.874
 

Minimal Detectable Change (MDC)

Chronic Pain: (calculated from Osman et al., 2000)

  • MDC for entire group (n = 60): 7.97

Cut-Off Scores

Chronic Pain: (Sullivan, 2009; n = 851; mean age = 42.4 years (SD not reported); mean duration work absence = 6.9 months (SD not reported); Canadian sample)

  • > 30 in total score indicates clinically relevant level of catastrophizing

Normative Data

Chronic Pain: (Sullivan, 2009)

  • Median total score: 20
  • 1st to 3rd quartile total score: 10-30

Test/Retest Reliability

Chronic Pain: (Xu et al., 2015; n = 55; mean age = 58.3 (13.6) years; mean pain duration = 21.1 (31.1) months; Chinese sample)

  • Excellent short-term test-retest reliability (5-7 days): (ICC = .94)

Chronic Pain: (Lamé et al., 2008; n = 48; mean age = 54.7 (13.1) years; Dutch sample)

  • Adequate long-term test-retest reliability (52 days): (ICC = .73)

Internal Consistency

Chronic Pain: (Osman et al., 2000)

  • Excellent for full scale: Cronbach's alpha for total score = .92
  • Adequate to Excellent for subscales: Cronbach's alpha for subscale scores = .75-.85

Criterion Validity (Predictive/Concurrent)

Concurrent Validity

Chronic Pain: (Osman et al., 2000)

  • Logistic regression analyses indicated that the PCS is useful for differentiating between chronic pain group and nonclinical community group (standardized estimate = .42, t = 4.99, p < .001)

Chronic Pain: (Osman et al., 1997; n = 174; mean age = 20.8 (4.1) years)

  • Hotelling’ T2 showed that the clinic seeking-treatment undergraduate students scored significantly higher than the non-clinic undergraduate students on the PCS total score (F(170) = 23.36, p < .001) and all subscale scores (F(170) = 12.06-27.71, p < .001)

Construct Validity

Chronic Pain: (Xu et al., 2015)

  • Excellent construct validity, both CFA and EFA confirmed a second-order three-factor structure (goodness of fit indexes: CMIN/DF = 1.68, NNFI = 0.948, CFI = 0.96, GFI = 0.907, RMSEA = 0.068)
  • Adequate correlation between PCS and HADS (r = .48)
  • Adequate correlation between PCS and BPI (r = .52)
  • Poor correlation between PCS and SF-12 (r = .24)

Content Validity

“The scores for each item were normally distributed, and none of the items showed a low item-total correlation (r < .40). Therefore, all the items were included in the Chinese version of the PCS.” (Xu et al., 2015, p. E1063)

Face Validity

Not statistically assessed, however, for the PCS-Simplified Chinese version, “the committee concluded that all the items could be employed to assess catastrophizing with good face validity, as demonstrated by other versions in different cultures” (Xu et al., 2015, p. E1061)

Healthy Adults

back to Populations

Standard Error of Measurement (SEM)

Nonclinical Undergraduate Students: (calculated from D'Eon et al., 2004; n = 505; mean age = 20.4 (3.9) years; Canadian sample)

  • SEM for entire group (n = 505): 3.15

Nonclinical Community Population: (Osman et al., 2000; n= 215; mean age = 35.1 (11.6) years)

  • SEM for entire group (n = 215): 2.26

Minimal Detectable Change (MDC)

Nonclinical Undergraduate Students: (calculated from D'Eon et al., 2004)

  • MDC for entire group (n = 505): 8.72

Nonclinical Community Population: (Osman et al., 2000)

  • MDC for entire group (n = 215): 6.27

Cut-Off Scores

Nonclinical Undergraduate Students: (Sullivan et al., 1995; n=438, Canadian sample)

  • Those scoring above 24 on the PCS were classified as catastrophizers

Normative Data

Nonclinical Undergraduate Students: (Sullivan et al., 1995; n=438, Canadian sample)

  • Study 1:
    • Women: (mean = 19.5, SD = 8.5)
    • Men: (mean = 16.4, SD = 7.3)
  • Study 2
    • Catastrophizer mean: 31.4-33.4 (SD 6.33-6.53)
    • Non-catastrophizer mean: 6.7-6.6 (SD 2.74-3.25)

Test/Retest Reliability

Nonclinical Undergraduate Students: (Sullivan et al., 1995; Canadian sample)

  • Excellent test-retest reliability across 6 weeks: (r = .75; n = 40; mean age = 18.8 (4.6) years)
  • Excellent test-retest reliability across 10 weeks: (r = .70; n = 60; mean age = 19.5 (5.8) years)

Interrater/Intrarater Reliability

Nonclinical Undergraduate Students:

  • Adequate interrater reliability between self-report and significant-other proxy report: (r = .35) (Cano et al., 2005)

Internal Consistency

Nonclinical Undergraduate Students: (Sullivan et al., 1995; n = 429; mean age = 20.1 (5.1) years; Canadian sample)

  • Excellent for full scale: Cronbach's alpha for total score = .87
  • Poor to Excellent for subscales:  Cronbach's alpha for subscale scores = .60 (magnification)-.87 (rumination)

Nonclinical Community Population: (Osman et al., 2000)

  • Excellent for full scale: Cronbach's alpha for total score = .95
  • Excellent for subscales: Cronbach's alpha for total score = .88-.95

Criterion Validity (Predictive/Concurrent)

Predictive validity:

Nonclinical Undergraduate Students: (Sullivan et al., 1995)

  • Adequate correlation between baseline PCS and pain rating during the ice water immersion at 10 weeks (r = .33)

Concurrent validity:

Nonclinical Undergraduate Students: (Osman et al., 1997; n = 220; mean age = 19.9 (3.4) years)

  • Adequate correlation between PCS and INTRP (r = .59)
  • The rumination subscale significantly predicted current pain severity (standardized β = .32, t = 3.04, p < .01)
  • Hotelling’ T2 showed that the clinic seeking-treatment undergraduate students scored significantly higher than the non-clinic undergraduate students on the PCS total score (F(170) = 23.36, p < .001) and all subscale scores (F(170) = 12.06-27.71, p < .001)

Nonclinical Community Population: (Osman et al., 2000)

  • Logistic regression analyses indicated that the PCS is useful for differentiating between chronic pain group and nonclinical community group (standardized estimate = .42, t = 4.99, p < .001)

Construct Validity

Nonclinical Undergraduate Students: (Sullivan et al., 1995)

  • Two-factor mixed ANOVA indicated a significant positive association between the PCS and catastrophic thoughts in response to the ice water immersion (F(1, 28) = 30.9, p < .001; post-hoc test t(28) = 5.4, p < .001)

Nonclinical Undergraduate Students: (Osman et al., 1997)

  • Excellent construct validity, CFA supported the second-order three-factor model (goodness of fit indexes: chi-squared = 113.24, df = 61, chi-squared/df = 1.86, NFI = .93, CFI = .97, RMSEA = .06)
  • Poor correlations between PCS and demographic variables (r = -.04-.19)
  • Poor correlation between PCS and MCSD (r = -.28)
  • Poor correlation between PCS and MASQ-Depression (r = .21)
  • Adequate correlation between PCS and MASQ-Anxiety (r = .37)

Nonclinical Community Population: (Osman et al., 2000)

  • Excellent construct validity, CFA supported a second-factor three-factor model (goodness of fit indexes: R-CFI = .97, NNFI = .97, RMSEA = .07)
  • Poor correlations between PCS and demographic variables (r = .01-.14)
  • Poor correlation between PCS and PANAS-PA (r = -.30)
  • Adequate correlation between PCS and PANAS-NA (r = .31)
  • Adequate correlations between PCS and pain indices (r = .51-.57)

Osteoarthritis

back to Populations

Standard Error of Measurement (SEM)

Knee Osteoarthritis: (Ong et al., 2020; n = 675; Mean Age = 65.52 years (SD = 6.84); PCS scores collected 1–2 weeks prior to knee replacement surgery; Singapore sample; Patient-Reported Outcome Measures (PROMs) self-administered in either English or Chinese)

  • SEM for entire group (n = 675): 2.584

Minimal Detectable Change (MDC)

Knee Osteoarthritis: (Ong et al., 2020)

  • MDC for entire group (n = 675): 7.16

Normative Data

Knee Osteoarthritis: (Ong et al., 2020)

  • Mean PCS score: 12.65 (SD = 10.55)

Table 1 Sociodemographic and clinical characteristics of patients with knee osteoarthritis

Characteristics

(N = 675)

Age, yearsa

66 (6.84)

   

Ethnic group

n (%)

    Chinese

564 (83.6)

    Malay

49 (7.3)

    Indian

48 (7.1)

    Others

14 (2.1)

   

Women, n (%)

475 (70.4)

BMI (kg/m2a

28.26 (5.02)

   

Duration of knee pain

n (%)

    < 1 year

81 (12.0)

    1–2 years

124 (18.4)

    3–5 years

195 (28.9)

    6–8 years

101 (15.0)

    9–12 years

84 (12.4)

    13–15 years

24 (3.6)

    > 15 years

66 (9.8)

   

Education

n (%)

    Primary or below

280 (41.5)

    Secondary

277 (41.0)

    Post-secondary and above

118 (17.5)

   

Type of arthroplasty

n (%)

    TKR

619 (91.7)

    UKA

56 (8.3)

   

PROMsb

 

    PCS total (0–52)

11 (4, 19)

    PCS rumination (0–16)

4 (2, 7)

    PCS magnification (0–12)

2 (0, 4)

    PCS helplessness (0–24)

4 (1, 9)

    SF-36 physical component summaryc

31.93 (26.19, 39.46)

    SF-36 mental component summaryc

53.93 (46.09, 60.98)

    WOMAC pain (0–100)

35 (25, 45)

    WOMAC function (0–100)

26.5 (17.6, 36.8)

    HADS anxiety (0–21)

3 (1, 6)

    HADS depression (0–21)

2 (1, 4)

    Pain self-efficacy (0–60)

39 (32, 47)

    Lubben Social Network Scale (0–60)

34 (27, 39)

TKR total knee replacement, UKA unicompartmental knee arthroplasty, PCS Pain Catastrophizing Scale, SF-36 short form 36 health survey, WOMAC The Western Ontario and McMaster Universities Osteoarthritis Index, HADS Hospital Anxiety and Depression Scale, PROMs patient-reported outcome measures, amean (standard deviation), bmedian (interquartile range), cSingapore norm-based scoring with mean (SD) 50 (10)

 

Internal Consistency

Knee Osteoarthritis: (Ong et al., 2020)

  • Excellent internal consistency for the PCS (Cronbach's alpha = 0.94*)

*Scores higher than 0.9 may indicate redundancy in the scale questions.

Construct Validity

Knee Osteoarthritis: (Ong et al., 2020)

  • The authors hypothesized that the PCS would
  • Construct validity was assessed by using 7 a priori hypotheses regarding how the PCS correlations with other instruments. In all instances, the authors’ hypotheses about the magnitude and direction of the correlation were fulfilled.

Construct Validities of PCS: Spearman’s rank correlations of PCS with other patient-reported outcome measures

PCS total score

Hypothesised correlations (Magnitude)

Hypothesised correlations (Direction)

Actual correlation (r)

Hypothesis met

Internal resource

 

 

 

 

    Pain self-efficacy questionnaire

Moderate

− 0.42*

Yes

         

Mental well-being

 

 

 

 

    HADS

Strong

+

0.58*

Yes

    SF-36 mental health

Moderate

− 0.34*

Yes

         

Physical well-being

 

 

 

 

    WOMAC physical function

Moderate

+

0.48*

Yes

    WOMAC pain

Moderate

+

0.47*

Yes

    SF-36 physical functioning

Moderate

− 0.31*

Yes

         

Social well-being

 

 

 

 

    Lubben Social Network Score

Weak

±

− 0.05

Yes

r = Spearman’s rho correlation coefficient, PCS Pain Catastrophizing Scale, SF-36 short form 36 health survey, WOMAC The Western Ontario and McMaster Universities Osteoarthritis Index, HADS Hospital Anxiety and Depression Scale

*p < 0.0071 (Bonferroni’s adjusted)

Content Validity

Knee Osteoarthritis: (Ong et al., 2020)

  • The authors determined the PCS had adequate content validity based on the cognitive debriefing interviews with 10 patients (Mean Age = 62 years) who were recruited prior to planned knee replacement for primary knee osteoarthritis.
  • All 10 patients completed the PCS in under 7 minutes (9 completed the questionnaire in under 5 minutes) and no questions were skipped.
  • During the cognitive debriefing interviews, patients reported the language was clear and content relevant to their pain experience.

Floor/Ceiling Effects

Knee Osteoarthritis: (Ong et al., 2020)

  • Adequate ceiling effect of 0.14% for the PCS
  • Adequate floor effect of 8.63% for the PCS

Mixed Populations

back to Populations

Normative Data

Mixed Population: (Wheeler et al., 2019; n = 42,976 clinical and nonclinical participants; Mean Age = 45 years (SD = 12); meta-analysis of 220 studies reporting baseline PCS in adults)

  • Ratio of males, females, and sex not reported = 55:32:13
  • Mean PCS score = 20.22 (weighted SD = 10.26, 95% CI = 19.30-21.14)

Test/Retest Reliability

Meta-Analysis: (Wheeler et al., 2019; test-retest reliability reported across 6 studies (n = 317); time between test and retest: range = 7–135 days)

  • Excellent test-retest reliability for PCS total score (Spearman ρ = 0.88, 95% CI = 0.83-0.93)

Internal Consistency

Meta-Analysis: (Wheeler et al., 2019; internal consistency reported across 40 studies for the PCS total score and 21 studies for the PCS subscales)

  • Excellent internal consistency for the PCS total score (Cronbach's alpha = 0.92*, 95% CI = 0.91-0.93)
  • Excellent internal consistency for the rumination subscale (Cronbach's alpha = 0.89, 95% CI = 0.87-0.91)
  • Adequate internal consistency for the magnification subscale (Cronbach's alpha = 0.77, 95% CI = 0.73-0.82)
  • Excellent internal consistency for the helplessness subscale (Cronbach's alpha = 0.88, 95% CI = 0.86-0.9)

*Scores higher than 0.9 may indicate redundancy in the scale questions

Bibliography

Cano, A., Leonard, M. T., & Franz, A. (2005). The significant other version of the Pain Catastrophizing Scale (PCS-S): Preliminary validation. Pain119(1-3), 26-37. https://doi.org/10.1016/j.pain.2005.09.009

D'Eon, J.L., Harris, C.A. & Ellis, J.A. (2004). Testing factorial validity and gender invariance of the Pain Catastrophizing Scale. Journal of Behavioral Medicine27(4), 361-372. https://doi.org/10.1023/B:JOBM.0000042410.34535.64

Lamé, I. E., Peters, M. L., Kessels, A. G., Van Kleef, M., & Patijn, J. (2008). Test-retest stability of the pain catastrophizing scale and the tampa scale for kinesiophobia in chronic pain over a longer period of time. Journal of Health Psychology13(6), 820-826. https://doi.org/10.1177/1359105308093866

Ong W.J., Kwan Y.H., et al. (2020). Measurement properties of Pain Catastrophizing Scale in patients with knee osteoarthritis. Clinical Rheumatology, Published online June 10, 2020. https://doi.org/10.1007/s10067-020-05163-8

Osman, A., Barrios, F. X., Gutierrez, P. M., Kopper, B. A., Merrifield, T., & Grittmann, L. (2000). The Pain Catastrophizing Scale: Further psychometric evaluation with adult samples. Journal of Behavioral Medicine23(4), 351-365. https://doi.org/10.1023/A:1005548801037

Osman, A., Barrios, F. X., Kopper, B. A., Hauptmann, W., Jones, J., & O'Neill, E. (1997). Factor structure, reliability, and validity of the Pain Catastrophizing Scale. Journal of Behavioral Medicine20(6), 589-605. https://doi.org/10.1023/A:1025570508954

Sullivan, M. J. (2009). The Pain Catastrophizing Scale user manual. https://sullivan-painresearch.mcgill.ca/pdf/pcs/PCSManual_English.pdf

Sullivan, M. J., Bishop, S. R., & Pivik, J. (1995). The pain catastrophizing scale: Development and validation. Psychological Assessment7(4), 524-532. https://doi.org/10.1037/1040-3590.7.4.524

Wheeler C.H.B., Williams A.C.C., et al. (2020). Meta-analysis of the psychometric properties of the Pain Catastrophizing Scale and associations with participant characteristics. Pain. 2019;160(9):1946-1953. https://doi.org/10.1097/j.pain.0000000000001494

Xu, X., Wei, X., Wang, F., Liu, J, & Chen, H. (2015). Validation of a simplified Chinese version of the pain catastrophizing scale and an exploration of the factors predicting catastrophizing in pain clinic patients. Pain Physician, 18, E1059-E1072.  https://www.painphysicianjournal.com/current/pdf?article=MjQ1NQ%3D%
3D&journal=92