Elsevier

The Spine Journal

Volume 10, Issue 6, June 2010, Pages 469-474
The Spine Journal

Clinical Study
Neck Disability Index, short form-36 physical component summary, and pain scales for neck and arm pain: the minimum clinically important difference and substantial clinical benefit after cervical spine fusion

https://doi.org/10.1016/j.spinee.2010.02.007Get rights and content

Abstract

Background Context

The Neck Disability Index (NDI), the short form-36 (SF-36) physical component summary (PCS), and pain scales for arm and neck pain are increasingly used to evaluate treatment effectiveness after cervical spine surgery. The minimum clinically important difference (MCID) is a threshold of improvement that is clinically relevant to the patient. However, the true goal is to provide the patient with a substantial clinical benefit (SCB).

Purpose

This study determines the MCID and SCB using common anchor-based methods for NDI, PCS, and pain scales for arm and neck pain in patients undergoing cervical spine fusion for degenerative disorders.

Study Design/Setting

The study setting is a longitudinal cohort in a multisurgeon spine specialty clinic.

Patient sample

The sample comprises 505 patients who underwent a cervical fusion for degenerative spine conditions and who have prospectively collected outcome scores with a minimum 1-year follow-up.

Outcome Measures

The outcome measures of the study were NDI, SF-36, and numeric rating scales for arm and neck pain.

Methods

The MCID and SCB values for NDI, PCS, and pain scales for arm and neck pain were determined using receiver operating characteristic (ROC) curve analysis with the Health Transition Item of the SF-36 as an anchor. The Health Transition Item asks a patient “Compared to one year ago, how would you rate your health in general now?” with answers ranging from “Much Better,” “Somewhat Better,” “About the Same,” “Somewhat Worse,” to “Much Worse.” An ROC curve was constructed for each measure. The ROC curve–derived MCID was the change score with equal sensitivity and specificity to distinguish the “Somewhat Better” from the “About the Same” patients. The ROC curve–derived SCB was the change score with equal sensitivity and specificity to distinguish the “Much Better” from the “Somewhat Better” patients. Distribution-based methods including the standard error of the mean and the minimum detectable change were also used to calculate MCID.

Results

The calculated MCID is 7.5 for the NDI, 4.1 for SF-36 PCS, and 2.5 for arm and neck pain. The calculated SCB is 9.5 for the NDI, 6.5 for SF-36 PCS, and 3.5 for arm and neck pain.

Conclusions

Patients with an eight-point decrease in NDI, a 4.1-point increase in PCS, and a three-point decrease in arm or neck pain can detect a minimally clinically important change. Patients with a 10-point decrease in NDI, a 6.5-point increase in PCS, and a four-point decrease in arm or neck pain can detect an SCB after cervical spine fusion.

Introduction

Evidence & Methods

Spinal fusion is a common procedure for a wide range of degenerative conditions. It is unclear whether outcome metrics primarily describing pain and pain-restricted function are adequate descriptors of outcome in this diverse group.

In a heterogeneous cohort of patients having cervical fusion for a degenerative process, the authors defined clinically important improvements in pain score, NDI, and the PCS (of the SF-36) by a post-hoc anchor-based method. Remarkably, less than half of the subjects reported any improvement in the post-hoc anchor: self-reported improvement in general health. MCID and SCB were calculated for various metrics using this anchor and appeared somewhat small. External metrics (eg, pain medication, health care utilization, occupational status) were not evaluated.

The authors' method does not appear to persuasively describe “good outcomes” in this setting. Unlike fusion in the lumbar spine, where pain is nearly always a central presenting feature, many cervical procedures are done to prevent or correct neurologic injury or deformity, independent of pain considerations. In addition, a patient's perception of “general health” after these procedures may also be independent of changes in neck discomfort. Separate analyses of specific conditions or rationale for surgery may show better validity for the post-hoc anchor-based calculations than this analysis by operation-type.

– The Editors

Patient-reported outcome measures or health-related quality of life surveys are increasingly becoming the standard measure for treatment effectiveness in spine surgery. Most studies report changes in group means before and after treatment. However, group means cannot be readily used in clinical practice to interpret changes on an individual basis. Previous authors presented the concept of minimum clinically important difference (MCID), “the smallest change that is important to patients” [1], [2], [3], [4], [5]. That is, the number of points a patient's score has to change for the patient to detect the smallest difference. A newer concept, the substantial clinical benefit (SCB), has been suggested by Glassman et al. [6] as a more realistic target value, arguing that the MCID is a floor value rather than a true goal of treatment. The SCB is the number of points a patient's score has to change for the patient to tell that he is much better.

Minimum clinically important difference values for the Neck Disability Index (NDI) [7] and pain scales after nonsurgical treatment have been reported by several authors [8]., [9], [10]. Cleland et al. [8] reported an MCID of 7.0 using an anchor-based method for NDI in a sample of 38 patients who underwent physical therapy for cervical radiculopathy. Pool et al. [9] reported a value of 3.5 using receiver operating characteristic (ROC) analysis in 183 patients with nonspecific neck pain undergoing physiotherapy or continued care from a general practitioner. Stratford et al. [10] using physician assessments as an anchor reported a value of 5.0 as an MCID for NDI in 48 patients presenting with neck pain seen at a physiotherapy clinic. There have been no studies that have reported MCID values for NDI in patients undergoing surgery for degenerative cervical spine disorders.

The purpose of this study was to calculate the MCID and SCB for the NDI, the Medical Outcomes short form-36 (SF-36) [11] physical component summary (PCS), and pain scales for arm and neck pain using common anchor-based (comparing outcome scores with another measurement) methods in patients with degenerative conditions of the cervical spine undergoing fusion.

Section snippets

Methods

Patients seen at a multisurgeon spine specialty practice undergoing cervical fusion for degenerative disorders, including spondylosis, disc herniation, and stenosis, from March 2000 to December 2006 who had complete preoperative and 1-year postoperative NDI [7], SF-36 [11], and numeric rating scales (0–10) for neck and arm pain [12] collected prospectively were included. All questionnaires were completed by the patients at home and were returned by mail.

The NDI is a 10-item self-administered

Results

Six hundred eighty-two consecutive patients had complete preoperative outcome measures. Of these, 505 (74%) completed outcome measures 1 year after surgery. The mean follow-up was 12.4±1.5 months. These patients had an average age of 52.6±10.2 years. Three hundred thirty-three (65.9%) were female and 88 (17.4%) were smokers. There was a similar distribution of fusion for spondylosis (170, 34%), herniated disc (152, 30%), and stenosis (183, 36%). There was no statistically significant difference

Discussion

Clinical outcome measures are increasingly used to monitor a patient's progress through different treatment modalities in spine surgery. Aside from published changes in group means when comparing different treatment modalities, the tools to interpret changes in outcome scores for a given individual in a clinical setting are lacking. What does a change in these different outcome scores mean? The most common metric used is the MCID, the smallest change in score that is clinically important to the

References (32)

  • D. Osoba et al.

    Interpreting the significance of changes in health-related quality-of-life scores

    J Clin Oncol

    (1998)
  • S.D. Glassman et al.

    Defining substantial clinical benefit following lumbar spine arthrodesis

    J Bone Joint Surg Am

    (2008)
  • H. Vernon et al.

    The Neck Disability Index: a study of reliability and validity

    J Manipulative Physiol Ther

    (1991 Sep)
  • J.A. Cleland et al.

    The reliability and construct validity of the Neck Disability Index and patient specific functional scale in patients with cervical radiculopathy

    Spine

    (2006)
  • J.J. Pool et al.

    Minimal clinically important change of the Neck Disability Index and the numerical rating scale for patients with neck pain

    Spine

    (2007)
  • P.W. Stratford et al.

    Using the Neck Disability Index to make decisions concerning individual patients

    Physiother Can

    (1999)
  • Cited by (310)

    • Diffusion basis spectrum imaging predicts long-term clinical outcomes following surgery in cervical spondylotic myelopathy

      2023, Spine Journal
      Citation Excerpt :

      Quality-of-life was measured by the SF-36 physical (PCS) and mental (MCS) component summary (MCIDs: 4 points) [22]. The neck disability index (NDI) was used to measure self-reported neck pain (MCID: 7.5 points) [23]. Long-term patient satisfaction was measured by the North American Spine Society (NASS) satisfaction index [24], where choice 1 was considered “improved” whereas choices 2, 3, or 4 were considered “not improved.”

    View all citing articles on Scopus

    FDA device/drug status: not applicable.

    Author disclosures: LYC (trips/travel, Medtronic Sofamor Danek); SDG (royalties, Medtronic; consulting, Medtronic; trips/travel, Medtronic; research: staff and/or materials, Medtronic, Norton Healthcare; grants, Norton Healthcare Community Trust Fund; fellowship support, Norton Healthcare); MJC (royalties, Medtronic Sofamor Danek; consulting, Medtronic Sofamor Danek; trips/travel, Medtronic Sofamor Danek; research: staff and/or materials, Medtronic Sofamor Danek, Norton Healthcare; fellowship support, Norton Healthcare); PAA (royalties, Stryker; stock ownership, including options and warrants, Pioneer; private investments, including venture capital, start-ups, Titan, Expanding Orthopedics; consulting, Pioneer, Medtronic; trips/travel, Medtronic; scientific advisory board, Pioneer; grants, NASS).

    View full text