The Psychometric Properties of the Neck Disability Index: a Comprehensive Review

The NDI has been used in a variety of populations, including mechanical neck pain (MNP),^1-8 cervical radiculopathy (CR),^9,10 whiplash associated disorder (WAD),^11,12 and mixed non-specific neck pain (Mixed-NSNP),^13-18 with a wide range of symptom durations (acute, subacute, and chronic). Vernon and Mior^19,20 suggest the following score interpretation: 0-4 represents no disability; 5-14 represents mild disability; 15-24 represents moderate disability; 25-34 represents severe disability (check out workers compensation law firm); and greater than 34 represents complete disability. However, Vernon and Mior¹⁹ did not describe the process for how these ratings were derived and, to date, no validation of these categories has been performed.²¹ The NDI has been found to possess excellent test-retest reliability, strong construct validity, strong internal consistency and good responsiveness in assessing disability in patients with mechanical neck pain (MNP),^1,2,5-8 cervical radiculopathy (CR),^9,10 whiplash associated disorder (WAD),^11,19,22 and mixed non-specific neck pain (Mixed-NSNP).^{13-15,17,18,23}

SEM, MDC and MCID

The standard error of measurement (SEM) has been found to be 3.0,⁶ 5.7,¹⁰ and 0.60¹⁷ – 1.80,¹⁸ in patients with MNP, CR, and Mixed-NSNP, respectively. In patients with MNP, the minimum detectable change (MDC) for the NDI has been reported to be 19.6 percentage points,² 8.4,⁶ 10.2,⁸ 4.2,¹⁸ and 5.²³ In patients with cervical radiculopathy, the MDC has been reported to be 13.4¹⁰ and 10.2.⁹ In patients with Mixed-NSNP, the MDC has been found to be 10.5,¹⁵ 1.66,¹⁷ and 4.20.¹⁸ In patients with MNP, the minimum clinically important difference (MCID) for the NDI has been found to be 19 percentage points,² 3.5,⁶ 7.5,⁸ and 5.²³ The MCID was found to be 8.5¹⁰ in patients with CR and 3.5¹⁵ in patients with Mixed-NSNP. Therefore, clinicians shoulder consider the minimum detectable change (MDC) for the NDI to approximate 4, 7, and 12 points (0-50) in patients with mixed non-specific neck pain, mechanical neck pain, and cervical radiculopathy, respectively. Similarly, the minimum clinically important difference (MCID) for the NDI should be considered to approximate 4, 6, and 9 points (0-50) in patients with Mixed-NSNP, MNP, and CR, respectively.

Responsiveness        

The NDI has been found to possess acceptable or satisfactory responsiveness (AUC > 0.70) in its ability to distinguish improved from stable patients in mechanical neck pain (AUC = 0.83²; 0.9²³; 0.75⁶; 0.79⁸), cervical radiculopathy (AUC = 0.74¹⁰; 0.57⁹), and Mixed-NSNP (Responsiveness ratio = 1.82¹⁷).

Test-retest reliability         

In patients with MNP, the NDI has demonstrated fair test-retest reliability (ICC = 0.50²), moderate test-retest reliability (ICC = 0.64⁸), and excellent test-retest reliability (ICC = 0.86⁶, r = 0.89¹⁹, r = 0.81 – 0.99¹, r = 0.94²³, r = 0.96⁷). In patients with CR, the NDI has demonstrated fair test-retest reliability (ICC = 0.55¹⁰; ICC = 0.68⁹). In patients with Mixed-NSNP, the NDI has shown excellent test-retest reliability (ICC = 0.93¹⁴; ICC = 0.90¹⁷; ICC = 0.89¹⁸).

Internal consistency

The NDI has demonstrated high internal consistency (Cronbach’s alpha > 0.7) in patients with MNP (Cronbach’s alpha = 0.92⁵, 0.89⁷, 0.72⁴, 0.77⁴), WAD (Cronbach’s alpha = 0.8¹⁹), and Mixed-NSNP (Cronbach’s alpha = 0.864¹⁴).

Construct validity    

The NDI possesses strong/excellent construct validity in patients with MNP when compared with the Global Rating of Change (r = 0.52⁸; r = 0.81⁷), the Neck Bournemouth Questionnaire (r = 0.80⁴ pretreatment; 0.77⁴ posttreatment), the Neck Pain and Disability Scale (r = 0.86³), and Problem Elicitation Technique (r = 0.62³). However, in patients with CR, the NDI has demonstrated poor construct validity.⁹ In patients with WAD, the NDI possesses strong construct validity when compared with the Northwick Park Questionnaire (r = 0.88¹¹) and only adequate construct validity when compared with Problem Elicitation Technique (r = 0.57¹¹). In patients with Mixed-NSNP, the NDI has been found to possess adequate construct validity when compared with the Mental Component Summary (MCS) score of the SF-36 (r = 0.47¹⁶) or the Physical Component Summary (PCS) score of the SF-36 (r = 0.53¹⁶), and strong/excellent construct validity when compared with the Patient-Specific Functional Scale (r = 0.73¹⁸ at admission; r = 0.81¹⁸ at discharge) or the Disabilities of the Arm, Shoulder & Hand (DASH) scale (r = 0.75¹³).

Content validity       

Of the eleven problems identified by most patients with MNP during problem elicitation technique (PET) interviews,³ six of these were found to not be included as items in the NDI; more specifically, En et al³ found the dimension of frustration, as well as other emotional and social functions, are not addressed in the NDI,³ suggesting limitations or gaps for the content validity of this outcome measure. In patients with WAD, and following PET interviews, the nine most commonly identified problems were ‘work for wages’, ‘fatigued during the day’, ‘participation in sports’, ‘depression’, ‘drive a car’, ‘socialize with friends’, ‘sleep through the night’, ‘frustration’, and ‘anger’; however, only three of these problems are specifically included in the NDI (work, driving, and sleeping).¹¹ Additionally, in patients with WAD, exploratory factor analysis found a 2-factor instrument: (1) pain intensity and interference with level of cognitive functioning, and (2) interference with level of physical functioning.¹² Likewise, Ailliet et al¹³ concluded the unidimensionality of the NDI could not be confirmed by clinicians and researchers with expertise in neck pain; that is, the NDI likely measures more than physical function. Moreover, expert clinicians and researchers suggested to fully capture the “burden of disease”, a number of items were likely missing from the NDI including: computer work, participation in sports, mobility, mental issues, dizziness, symptom radiation, the distinction between household work, and professional activities.¹³ Ailliet et al¹³ also suggested that the NDI does not take into account the use of over-the-counter or prescription medication and its potential effect on modulating neck pain and disability.¹³

Over-the-counter or prescription medication use, participation in sports, and personal computer work are not addressed by the NDI suggesting the instrument may not completely capture important aspects of the modern disablement process in patients with neck pain.¹³ Missing data frequently occurs for item 8 (i.e. driving) in cultures that restrict females from driving;⁷ therefore, like the Neck Pain and Disability Scale (NPDS), it has been suggested that this item be answered by either a driver or passenger.⁷ The NDI has no “response set bias” suggesting subjects are responding to the content, not the item order or the sequencing of the responses format.⁵ The NDI does not have any serious floor or ceiling effects;^7,16 therefore, the NDI is able to assess the full range of disability in relation to neck pain. However, a clear differential item functioning (DIF) for the “headaches” item has been observed; moreover, women appear to score significantly worse than men with similar levels of disability.¹³

Unidimensionality  

The NDI has traditionally been accepted to be unidimensional^20,21 and tends to capture more of the physical aspects of pain and disability rather than the emotional and social aspects. However, although exploratory factor analyses had originally identified a 1-factor model that accounted for 59% to 65% of the variance,⁵ a 2-factor structure (1 subscale describing ADLs and the other subscale representing pain and concentration) recently explained 67.58% of the total variance.⁷ Additionally, another recent study found two factors to explain 53.8% of the variance within the NDI.¹³ Nevertheless, although the NDI adequately captures the physical aspects of neck pain, it may not capture the broader emotional (frustration, depression, and anger) and social function dimensions, which are frequently reported by patients with neck pain.^3,11

Summary of the psychometric properties of the NDI     

The NDI is the oldest and most widely used instrument for self-reporting of disability in patients with neck pain.^20,21,24 Its psychometric properties have been well established in a large number of independent studies;^20,21,24 furthermore, the NDI is the scale most widely validated among different populations and settings of patients with neck pain.²¹ The NDI has been found to possess fair,^2,9,10 moderate⁸ and excellent^{1,6,7,14,17-19,23} test-retest reliability, strong^{3,4,7,11,13,18} construct validity, strong^4,5,7,14,19 internal consistency and acceptable/satisfactory^{2,6,8,10,17,23} responsiveness in patients with mechanical neck pain, whiplash associated disorder, or mixed non-specific neck pain; additionally, in the main, it has been found to possess a 1-factor structure (i.e. unidimensional) for “physical disability”.^20,21

AUTHOR:

James Dunning, DPT, MSc (Manip Ther), MMACP (UK), FAAOMPT
Director, AAMT Fellowship in Orthopaedic Manual Physical Therapy
President, Alabama Physical Therapy & Acupuncture, Montgomery, AL

REFERENCES

1. Ackelman BH, Lindgren U. Validity and reliability of a modified version of the neck disability index. J Rehabil Med. Nov 2002;34(6):284-287.
2. Cleland JA, Childs JD, Whitman JM. Psychometric properties of the Neck Disability Index and Numeric Pain Rating Scale in patients with mechanical neck pain. Arch Phys Med Rehabil. Jan 2008;89(1):69-74.
3. En MC, Clair DA, Edmondston SJ. Validity of the Neck Disability Index and Neck Pain and Disability Scale for measuring disability associated with chronic, non-traumatic neck pain. Man Ther. Aug 2009;14(4):433-438.
4. Gay RE, Madson TJ, Cieslak KR. Comparison of the Neck Disability Index and the Neck Bournemouth Questionnaire in a sample of patients with chronic uncomplicated neck pain. J Manipulative Physiol Ther. May 2007;30(4):259-262.
5. Hains F, Waalen J, Mior S. Psychometric properties of the neck disability index. J Manipulative Physiol Ther. Feb 1998;21(2):75-80.
6. Jorritsma W, Dijkstra PU, de Vries GE, Geertzen JH, Reneman MF. Detecting relevant changes and responsiveness of Neck Pain and Disability Scale and Neck Disability Index. Eur Spine J. Dec 2012;21(12):2550-2557.
7. Shaheen AA, Omar MT, Vernon H. Cross-cultural adaptation, reliability, and validity of the Arabic version of neck disability index in patients with neck pain. Spine (Phila Pa 1976). May 1 2013;38(10):E609-615.
8. Young BA, Walker MJ, Strunce JB, Boyles RE, Whitman JM, Childs JD. Responsiveness of the Neck Disability Index in patients with mechanical neck disorders. Spine J. Oct 2009;9(10):802-808.
9. Cleland JA, Fritz JM, Whitman JM, Palmer JA. The reliability and construct validity of the Neck Disability Index and patient specific functional scale in patients with cervical radiculopathy. Spine (Phila Pa 1976). Mar 1 2006;31(5):598-602.
10. Young IA, Cleland JA, Michener LA, Brown C. Reliability, construct validity, and responsiveness of the neck disability index, patient-specific functional scale, and numeric pain rating scale in patients with cervical radiculopathy. Am J Phys Med Rehabil. Oct 2010;89(10):831-839.
11. Hoving JL, O’Leary EF, Niere KR, Green S, Buchbinder R. Validity of the neck disability index, Northwick Park neck pain questionnaire, and problem elicitation technique for measuring disability associated with whiplash-associated disorders. Pain. Apr 2003;102(3):273-281.
12. Nieto R, Miro J, Huguet A. Disability in subacute whiplash patients: usefulness of the neck disability index. Spine (Phila Pa 1976). Aug 15 2008;33(18):E630-635.
13. Ailliet L, Knol DL, Rubinstein SM, de Vet HC, van Tulder MW, Terwee CB. Definition of the construct to be measured is a prerequisite for the assessment of validity. The Neck Disability Index as an example. J Clin Epidemiol. Jul 2013;66(7):775-782; quiz 782 e771-772.
14. McCarthy MJ, Grevitt MP, Silcocks P, Hobbs G. The reliability of the Vernon and Mior neck disability index, and its validity compared with the short form-36 health survey questionnaire. Eur Spine J. Dec 2007;16(12):2111-2117.
15. Pool JJ, Ostelo RW, Hoving JL, Bouter LM, de Vet HC. Minimal clinically important change of the Neck Disability Index and the Numerical Rating Scale for patients with neck pain. Spine (Phila Pa 1976). Dec 15 2007;32(26):3047-3051.
16. Riddle DL, Stratford PW. Use of generic versus region-specific functional status measures on patients with cervical spine disorders. Phys Ther. Sep 1998;78(9):951-963.
17. Vos CJ, Verhagen AP, Koes BW. Reliability and responsiveness of the Dutch version of the Neck Disability Index in patients with acute neck pain in general practice. Eur Spine J. Nov 2006;15(11):1729-1736.
18. Westaway MD, Stratford PW, Binkley JM. The patient-specific functional scale: validation of its use in persons with neck dysfunction. J Orthop Sports Phys Ther. May 1998;27(5):331-338.
19. Vernon H, Mior S. The Neck Disability Index: a study of reliability and validity. J Manipulative Physiol Ther. Sep 1991;14(7):409-415.
20. Vernon H. The Neck Disability Index: state-of-the-art, 1991-2008. J Manipulative Physiol Ther. Sep 2008;31(7):491-502.
21. MacDermid JC, Walton DM, Avery S, et al. Measurement properties of the neck disability index: a systematic review. J Orthop Sports Phys Ther. May 2009;39(5):400-417.
22. Miettinen T, Leino E, Airaksinen O, Lindgren KA. The possibility to use simple validated questionnaires to predict long-term health problems after whiplash injury. Spine (Phila Pa 1976). Feb 1 2004;29(3):E47-51.
23. Stratford PW, Riddle DL, Binkley JM, Spadoni G, Westaway MD, Padfield B. Using the Neck Disability Index to make decisions concerning individual patients. Physiotherapy Canada. 1999;51:107-112.
24. Pietrobon R, Coeytaux RR, Carey TS, Richardson WJ, DeVellis RF. Standard scales for measurement of functional outcome for cervical pain or dysfunction: a systematic review. Spine (Phila Pa 1976). Mar 1 2002;27(5):515-522.