A prognostic approach to defining chronic pain

Michael Von Korff, Diana L Miglioretti

Research output: Contribution to journalArticle

122 Citations (Scopus)

Abstract

This paper develops a prognostic approach to defining chronic back pain. Possible and probable chronic back pain were defined, respectively, by a 50% and an 80% (or greater) probability of future clinically significant back pain. We assessed whether an empirically derived chronic pain classification satisfied these validating criteria among 1213 primary care back pain patients assessed at baseline and at 1, 2 and 5 year follow-ups. From multiple measures of back pain intensity and dysfunction, Latent Transition Regression Analysis empirically identified four pain severity latent classes: no pain; mild pain; moderate pain and limitation; and severe, limiting pain. From one observation point to the next, patients were most likely to remain in the same pain severity class, but chronic pain was better characterized as a dynamic state than a static trait. Among persons with severe, limiting pain, prognostic variables (depression, diffuse pain, pain persistence) improved prediction of future severe, limiting pain. A risk score developed from pain severity and prognostic measures identified risk levels corresponding to 50 and 80% probability thresholds for predicting future clinically significant back pain. At baseline and 1 year, 6.1 and 4.4% of study patients met or exceeded the 80% risk threshold for probable chronic back pain. An additional 20.3% at baseline and 12.5% at 1 year met or exceeded the 50% risk threshold for possible chronic back pain. Defining chronic pain prospectively, by risk thresholds for future clinically significant pain, provides an empirically grounded approach to chronic pain assessment.

Original languageEnglish (US)
Pages (from-to)304-313
Number of pages10
JournalPain
Volume117
Issue number3
DOIs
StatePublished - Oct 2005
Externally publishedYes

Fingerprint

Chronic Pain
Back Pain
Pain
Pain Measurement
Primary Health Care
Regression Analysis
Observation
Depression

ASJC Scopus subject areas

  • Clinical Neurology
  • Psychiatry and Mental health
  • Neurology
  • Neuroscience(all)
  • Pharmacology
  • Clinical Psychology

Cite this

A prognostic approach to defining chronic pain. / Von Korff, Michael; Miglioretti, Diana L.

In: Pain, Vol. 117, No. 3, 10.2005, p. 304-313.

Research output: Contribution to journalArticle

Von Korff, Michael ; Miglioretti, Diana L. / A prognostic approach to defining chronic pain. In: Pain. 2005 ; Vol. 117, No. 3. pp. 304-313.
@article{59caa963b16e4f989af0acbca1a94b73,
title = "A prognostic approach to defining chronic pain",
abstract = "This paper develops a prognostic approach to defining chronic back pain. Possible and probable chronic back pain were defined, respectively, by a 50{\%} and an 80{\%} (or greater) probability of future clinically significant back pain. We assessed whether an empirically derived chronic pain classification satisfied these validating criteria among 1213 primary care back pain patients assessed at baseline and at 1, 2 and 5 year follow-ups. From multiple measures of back pain intensity and dysfunction, Latent Transition Regression Analysis empirically identified four pain severity latent classes: no pain; mild pain; moderate pain and limitation; and severe, limiting pain. From one observation point to the next, patients were most likely to remain in the same pain severity class, but chronic pain was better characterized as a dynamic state than a static trait. Among persons with severe, limiting pain, prognostic variables (depression, diffuse pain, pain persistence) improved prediction of future severe, limiting pain. A risk score developed from pain severity and prognostic measures identified risk levels corresponding to 50 and 80{\%} probability thresholds for predicting future clinically significant back pain. At baseline and 1 year, 6.1 and 4.4{\%} of study patients met or exceeded the 80{\%} risk threshold for probable chronic back pain. An additional 20.3{\%} at baseline and 12.5{\%} at 1 year met or exceeded the 50{\%} risk threshold for possible chronic back pain. Defining chronic pain prospectively, by risk thresholds for future clinically significant pain, provides an empirically grounded approach to chronic pain assessment.",
author = "{Von Korff}, Michael and Miglioretti, {Diana L}",
year = "2005",
month = "10",
doi = "10.1016/j.pain.2005.06.017",
language = "English (US)",
volume = "117",
pages = "304--313",
journal = "Pain",
issn = "0304-3959",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - A prognostic approach to defining chronic pain

AU - Von Korff, Michael

AU - Miglioretti, Diana L

PY - 2005/10

Y1 - 2005/10

N2 - This paper develops a prognostic approach to defining chronic back pain. Possible and probable chronic back pain were defined, respectively, by a 50% and an 80% (or greater) probability of future clinically significant back pain. We assessed whether an empirically derived chronic pain classification satisfied these validating criteria among 1213 primary care back pain patients assessed at baseline and at 1, 2 and 5 year follow-ups. From multiple measures of back pain intensity and dysfunction, Latent Transition Regression Analysis empirically identified four pain severity latent classes: no pain; mild pain; moderate pain and limitation; and severe, limiting pain. From one observation point to the next, patients were most likely to remain in the same pain severity class, but chronic pain was better characterized as a dynamic state than a static trait. Among persons with severe, limiting pain, prognostic variables (depression, diffuse pain, pain persistence) improved prediction of future severe, limiting pain. A risk score developed from pain severity and prognostic measures identified risk levels corresponding to 50 and 80% probability thresholds for predicting future clinically significant back pain. At baseline and 1 year, 6.1 and 4.4% of study patients met or exceeded the 80% risk threshold for probable chronic back pain. An additional 20.3% at baseline and 12.5% at 1 year met or exceeded the 50% risk threshold for possible chronic back pain. Defining chronic pain prospectively, by risk thresholds for future clinically significant pain, provides an empirically grounded approach to chronic pain assessment.

AB - This paper develops a prognostic approach to defining chronic back pain. Possible and probable chronic back pain were defined, respectively, by a 50% and an 80% (or greater) probability of future clinically significant back pain. We assessed whether an empirically derived chronic pain classification satisfied these validating criteria among 1213 primary care back pain patients assessed at baseline and at 1, 2 and 5 year follow-ups. From multiple measures of back pain intensity and dysfunction, Latent Transition Regression Analysis empirically identified four pain severity latent classes: no pain; mild pain; moderate pain and limitation; and severe, limiting pain. From one observation point to the next, patients were most likely to remain in the same pain severity class, but chronic pain was better characterized as a dynamic state than a static trait. Among persons with severe, limiting pain, prognostic variables (depression, diffuse pain, pain persistence) improved prediction of future severe, limiting pain. A risk score developed from pain severity and prognostic measures identified risk levels corresponding to 50 and 80% probability thresholds for predicting future clinically significant back pain. At baseline and 1 year, 6.1 and 4.4% of study patients met or exceeded the 80% risk threshold for probable chronic back pain. An additional 20.3% at baseline and 12.5% at 1 year met or exceeded the 50% risk threshold for possible chronic back pain. Defining chronic pain prospectively, by risk thresholds for future clinically significant pain, provides an empirically grounded approach to chronic pain assessment.

UR - http://www.scopus.com/inward/record.url?scp=25644455075&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=25644455075&partnerID=8YFLogxK

U2 - 10.1016/j.pain.2005.06.017

DO - 10.1016/j.pain.2005.06.017

M3 - Article

C2 - 16153772

AN - SCOPUS:25644455075

VL - 117

SP - 304

EP - 313

JO - Pain

JF - Pain

SN - 0304-3959

IS - 3

ER -