The CNFDS is designed to evaluate the disability experienced by patients with CSM. It consists of three sections including 15 items that evaluate: the impact of neck pain including the patient's perception of the future impact of neck pain (items 1, 5, and 15), disability during everyday activities (items 2, 3, 4, 5, 7, 8, 9, 10, and 12), and social interactions and recreation (items 6, 9, 11, 13, and 14) [621]. Each item has three possible response categories (yes=0, occasionally=1, and no=2) for the first five items and the remaining items are scored in reverse (yes=2, occasionally=1, and no=0). The total score is 0-30. A higher score indicates greater disability (Appendix 1).

We asked for permission to translate the CNFDS into Persian. Then, the "forward-backward" procedure was applied to translate the CNFDS from English into Persian. Two general practitioners translated the questionnaire into Persian, and these were back translated into English by a health professional and a professional translator. A few changes were made after a careful review, and the provisional Persian version of the questionnaire was pilot tested to establish that it could be understood and that the questions measured what they were intended to measure [222324].

The final draft of the Iranian version was administered to a sample of patients newly diagnosed with CSM who were attending a neurosurgery clinic of a Shahid Beheshti University of Medical Sciences (Tehran, Iran) during April 2007 to June 2013. No restrictions were put on patient selection with regard to type of CSM, age, or other characteristics. The exclusion criteria were prior cervical spine surgery and spinal anomalies. The stenotic level(s) were localized on magnetic resonance or computed tomography images. All patients had typical symptoms of CSM and were surgical candidates. Two neurosurgical health professionals independently scored the questionnaires of patients admitted for surgery. Both independent observers assessed the patients on the same day. Scoring was performed on the day before surgery. The observers were unaware of the purpose of the study. Patients were assessed pre and postoperatively at the 6 months follow-up.

The key to treating CSM is to remove pressure from the spinal cord. Surgical methods to decompress the spinal cord include three approaches: (1) anterior cervical corpectomy from the front of the neck, as with anterior cervical discectomy and fusion and (2) cervical laminectomy and fusion and cervical laminoplasty from the back of the neck, and (3) combined procedure [25].

In addition to the CNFDS, the Iranian version of the mJOA was administered simultaneously to patients. The mJOA is a self-administered, disease-specific tool that originated from JOA score [1011]. It consists of four sections including 22 items: Motor dysfunction of the upper extremities (six items), motor dysfunction of the lower extremities (eight items), sensation (four items), and sphincter dysfunction (four items). The scores for each section are 0-5, 0-7, 0-3, and 0-3, respectively, giving a total score of 0-18. Higher score indicate less disability. In this study, the total mJOA score was used for the assessment.

The psychometric properties of the CNFDS were evaluated for reliability, validity, and responsiveness to change. (1) Reliability: Cronbach α coefficient was used to test reliability and internal consistency of the questionnaire, and α≥0.70 was considered satisfactory. In addition, we assessed test-retest reliability using the intraclass correlation coefficient (ICC) analysis. ICC values >0.80 were considered excellent stability [26]. (2) Validity: Construct validity was assessed with the item-scale correlations. Correlations were calculated using Pearson correlation coefficient (r) analysis. We expected that item scores would correlate higher with the hypothesized scale than the other scales. Correlation values ≥0.40 were considered satisfactory (r≥0.81-1.0, excellent; 0.61-0.80, very good; 0.41-0.60, good; 0.21-0.40, fair; and 0-0.20, poor) [19]. In addition the correlation between CNFDS and mJOA scores was assessed using Pearson correlation coefficient analysis to assess convergent validity (criterion validity). This type of validity is useful for showing that the construct of an instrument is meaningful. Values ≥0.40 were considered satisfactory (r≥0.81-1.0, excellent; 0.61-0.80, very good; 0.41-0.60, good; 0.21-0.40, fair; and 0-0.20, poor) [26]. (3) Responsiveness to change: Responsiveness was evaluated as a psychometric property of the questionnaire. As such, the patient's pre and postoperative scores were compared using the paired t-test to examine whether the CNFDS could capture the changes after surgery.

The Ethics Committee of Shahid Beheshti University of Medical Sciences (Tehran, Iran) approved this study.

Ninety-three patients with CSM were studied. The mean age of the patients was 54.3 (standard deviation [SD], 8.9) years, most were married (72%), and had completed primary or secondary education (61.3%). The patient's characteristics and their mJOA scores are shown in Table 1. Most patients with CSM had a developmentally narrow spinal canal, and the decompressive laminae were distributed from the C2 to T1 levels. The number of decompressed lamina was 3.1 (SD, 1.0). Most patients with CHD had a one- or two level discectomy distributed from the C2 to C7 levels. Significant differences were observed between the pre and postoperative assessments, indicating improvements on the outcomes and functionality on all subscales (p<0.001). However, no differences were detected between patients with cervical spinal stenosis and CHD (p≥0.05).

(1) The internal consistency for the CNFDS and its three sections as calculated by the Cronbach α coefficient is shown in Table 2. All sections exceeded the minimum reliability standard of 0.70 at the preoperative assessment. The Cronbach α for the scale was 0.84, indicating satisfactory results. The results are shown in Table 2. (2) Test-retest reliability: The ICC was excellent (0.95) for the CNFDS (95% confidence interval, 0.92-0.98) (Table 3).

Construct validity of the CNFDS was examined using item-scale correlations and criterion validity. The item scale correlation matrix between each item and the three CNFDS subscales are shown in Table 4. All correlations between items and the hypothesized scale were satisfactory, suggesting that the items had a substantial association with their own subscale. Pearson correlation coefficient exceeded 0.40 (range, 0.57 [Q10]-0.71 [Q4]).

Total score on the CNFDS was correlated strongly with total score on the mJOA, supporting the good convergent validity (r=-0.80, p<0.001).

Responsiveness to change was evaluated using the paired t-test. In all instances, the CNFDS detected changes after the intervention (surgery), indicating improvements in all subscales as expected. The outcomes are shown in Table 5.