As shown in table 2, no significant differences were observed between the groups of male and female patients with ischemic stroke in terms of their stroke risk factors, such as a previous history of stroke, a family history of stroke, transient ischemic attacks (TIA), hyperlipidemia, ischemic heart diseases and atrial fibrillation (p>0.05). Nevertheless, Pearson’s Chi-squared test showed a significant difference between the groups of male and female patients in terms of obesity as signified by a BMI equal/above 30 (χ²=24.93, df=1, p<0.0001), hypertension (χ²=5.49, df=1, p=0.019) and diabetes mellitus (χ²=7.47, df=1 , p=0.006). This means that compared to men, a higher percentage of women suffered from obesity (34.3% vs. 6.5%), hypertension (80.8% vs. 66.4%), and diabetes mellitus (50.5% vs. 31.8%).

The Cronbach's alpha coefficients obtained for determining the internal consistency reliability of the NIHSS were 0.881, 0.919 and 0.893 for the groups of male and female patients and the entire group of patients respectively. With regard to sample size and based on Streiner and Norman’s comments, these  coefficients are acceptable  (23).  Salter et al. believe Cronbach's alpha coefficient equal to or above 0.8 to be excellent, especially for tools assessing rehabilitation factors among stroke patients (24). The calculated Cronbach's alpha values for the groups of male and female patients and the entire group of patients with ischemic stroke thus indicated a good internal consistency reliability of NIHSS.

To assess the concurrent validity of the NIHSS in the groups of male and female patients with ischemic stroke, its correlation with the Barthel index was first assessed separately in which the results demonstrated a negative and significant correlation with the Barthel index in both groups of male patients (r= -0.43 and p<0.0001) and female patients (r= -0.63 and p<0.0001). The negative correlation was attributed to how, higher scores on the NIHSS meant greater neurological damage and higher scores on the Barthel index showed a lower functional disability. Fisher’s Z test was then used to find the significance of the difference between the correlation coefficients obtained in the groups of male and female patients, in which values of r₁ and r₂ were converted into Z_r1 and Z_r2 using Fisher’s table (22) and the data obtained were then inserted in the below equation and the final result was thus obtained as follows:

Fisher’s Z test showed significant differences between the correlation coefficients obtained in the groups of male and female patients with ischemic stroke at the significance level of p<0.05 and using the two-tailed test (Z=-1.99 and p=0.046). These findings clear that the correlation between the NIHSS scores and the Barthel index is stronger in the group of female patients compared to male patients group. The bivariate regression analysis was performed to determine what scores in the NIHSS (as the predictor variable) can explain variances in the Barthel Index scores (as the criterion variable) in the groups of male and female patients with ischemic stroke separately. Figure 1 represents the results obtained from the bivariate regression analysis and the difference in the slope of the line in two groups of male and female patients.

Figure 1: The slope of the line and the prediction equations for the Barthel scores from the NIHSS scores in two groups of male (n=107) and female (n=99) patients

The results of the bivariate regression analysis showed that NIHSS scores can significantly explain the variance of the Barthel Index scores in both groups of patients with ischemic stroke (p<0.0001). According to figure 1, the  coefficients of determination obtained in the two groups show that the NIHSS is a stronger predictor of functional disability as measured by the Barthel Index in female (R²=0.40) than in male (R²=0.18) patients.

Discussion

The NIHSS is known as a valid scale for predicting the amount of brain lesion and can also be used to assess the severity of the stroke. Nevertheless, there is no evidence on the applicability and other psychometric features of this tool among Iranian patients, especially with respect to gender differences. The present study used the NIHSS to assess post-stroke neurological deficits and to also determine the internal consistency reliability and concurrent validity of this tool in male and female groups of patients.

The descriptive results obtained through comparing the two gender groups showed no significant differences between them in terms of age or stroke risk factors such as a previous history of stroke, a family history of stroke, Transient Ischemic Attacks (TIA), hyperlipidemia, ischemic heart diseases and atrial fibrillation; the two groups were therefore matching in the majority of the risk factors. In contrast the female patients examined, experienced more severe neurological deficits and a greater functional disability after the stroke compared to the male patients. The differences observed between the two groups in the concurrent validity coefficients obtained may be partly due to the greater vulnerability of women to certain stroke risk factors such as obesity, hypertension and diabetes mellitus (Table 2).

The Cronbach's alpha coefficients obtained for the groups of male and female patients and also the entire group of patients with ischemic stroke (0.881, 0.919, and 0.893, in respective order) suggests that the Persian version of the NIHSS performed by a neurologist has good internal consistency reliability for the all sample population. These results  show  that  each  of  the  NIHSS items

have the required level of congruence and uniformity to assess the general concept of neurological deficits. In other words, each of the items measures the same structure and no redundancy is observed in them. Although the lack of similar studies on the reliability of the NIHSS by gender and also only a few studies have examined the internal consistency reliability of this scale, the review of these studies revealed the agreement with the present study. In one study, Sun et al. assessed the validity and reliability of the Chinese version of the NIHSS in 48 patients with ischemic stroke and found a Cronbach's alpha value of 0.92 for the scale, which agrees closely with the value found in the present study (2).

The concurrent validity of the NIHSS showed a negative and significant correlation with the Barthel Index in both the groups of male (r=-0.43) and female (r=-0.63) patients. Fisher’s Z test showed a significant difference between the correlation coefficients obtained in the two groups (p<0.05), which was much stronger in the female group compared to in the male group. The results of the bivariate regression analysis showed that NIHSS scores can predict Barthel Index scores in groups of male and female patients with ischemic stroke; however, its ability to predict functional disability was found to be stronger in the female group compared to in the male group. The NIHSS thus has a greater concurrent validity in the group of women than in the group of men; nonetheless, it can generally be used as a valid and efficient tool to predict a moderate level of functional disability as a major consequence of stroke. In line with these results, Frankel et al. showed that patients with more severe neurological deficits (as assessed by the NIHSS) have a worse prognosis and more likely identified as

a subset of patients with high degrees of poor outcomes in the first week following their acute stroke (25). Sun et al. assessed the concurrent validity of the NIHSS and found a negative significant correlation between this scale and the Barthel Index (r=-0.683) (2). Meyer et al. assessed the concurrent validity of the NIHSS through determining its correlation with the Barthel Index, which was found to be weak but significant association (r=-0.165) (26). Also Leyden et al. estimated this correlation=-0.48 immediately after the stroke and=-0.58 two hours after the incident (27), which is closely in line with the correlation obtained in the present study. Brott et al. confirmed the concurrent validity of the NIHSS through determining its correlation with the volume of cerebral infarction in 65 stroke patients by computed tomography (r=0.74) (6). Lyden et al. and also Saver et al. conducted similar studies to assess the concurrent validity of the NIHSS and reported the correlation between the scores of  this scale and the volume of cerebral infarction in stroke patients as r=0.37 and r=0.54, respectively (28, 29).

In line with the present research, previous studies conducted with different methods also emphasized the accuracy of the results obtained through the NIHSS and confirmed its concurrent validity. However, no studies were found that had performed this assessment from a gender-based perspective. The differences observed in both the stroke risk factors (table 2) and in the concurrent validity coefficients obtained in the groups of male and female patients may therefore be explained through the differences in the severity of neurological deficits (as indicated by the NIHSS scores) in the two groups (table 1); since it has already been proved how mild neurological deficits measured by the NIHSS

can reduce the concurrent validity of the scale in relation to the Barthel Index (25).

The limitations of the present study included the unavailability of records of NIHSS scores obtained by the patients upon admission. Moreover, the present study was conducted solely on patients with ischemic stroke and its results cannot be generalized to patients with the other types of stroke; however, these findings can be used as a basis for future comparisons in Iranian patients. The inter-rater reliability of the scale should be assessed in future studies and a comparison be made between the NIHSS at the times of admission and discharge.

Conclusion

Overall, the results of the present research confirmed that the Persian version of the NIHSS is a valid and reliable tool that applies to both male and female patients with ischemic stroke; nevertheless, it should be noted that the concurrent validity of the scale is stronger in the group of female than male patients.

Acknowledgements

The present study was an approved research project by Guilan University of Medical Sciences (GUMS). Hereby, the authors would like to express their gratitude to the Research and technology vice-chancellorship of GUMS. Also we would like to thank the colleagues in the clinical researches and development center of Poursina Hospital for their support for sampling from the stroke patients.

Conflict of Interest

The authors have no conflict of interest.

References

1. Jensen MB, Lyden P. Stroke scales: An update. Stroke Clinical Updates. 2006; 16:1-7.
2. Sun TK, Chiu SC, Yeh SH, Chang KC. Assessing Reliability and Validity of the Chinese Version of the Stroke Scale: Scale Development. Int J Nurs Stud 2006; 43(4):457-63.
3. Ingram SR, Sedlak SK. New Age in Stroke Treatment: the Role of Fibrinolytictherapy. Am J Nurs 2002; Suppl:17-21.
4. Spilker J, Kongable G, Barch C, Braimah J, Brattina P, Daley S, et al. Using the NIH Stroke Scale to Assess Stroke Patients. The NINDS rt-PA Stroke Study Group. J Neurosci Nurs 1997; 29(6):384-92.
5. Schlegel D, Kolb SJ, Luciano JM, Tovar JM, Cucchiara BL, Liebeskind DS, et al. Utility of the NIH Stroke Scale as a Predictor of Hospital Disposition. Stroke 2003; 34(1):134-7.
6. Brott T, Marler JR, Olinger CP, Adams HP Jr, Tomsick T, Barsan WG, et al. Measurements of Acute Cerebral Infarction: Lesion Size by Computed Tomography. Stroke 1989; 20(7):871-5.
7. Schiemanck SK, Post MW, Witkamp TD, Kappelle LJ, Prevo AJ. Relationship between Ischemic Lesion Volume and Functional Status in the 2nd Week after Middle Cerebral Artery Stroke. Neurorehabil Neural Repair 2005; 19(2):133-8.
8. Goldstein LB, Bertels C, Davis JN. Interrater Reliability of the NIH Stroke Scale. Arch Neural 1989; 46(6):660-2.
9. Williams LS, Yilmaz EY, Lopez-Yunez AM. Retrospective Assessment of Initial Stroke Severity with the NIH Stroke Scale. Stroke 2000; 31(4):858-62.
10. Sandelowski M. The Problem of Rigor in Qualitative Research. ANS Adv Nurs Sci 1986;8(3):27-37.
11. Granger CV, Dewis LS, Peters NC, Sherwood CC, Barrett JE. Stroke Rehabilitation: Analysis of Repeated Barthel Index Measures. Arch Phys Med Rehabil 1979; 60(1):14-7.
12. Oveisgharan S, Shirani S, Ghorbani A, Soltanzade A, Baghaei A, Hosseini S, Sarrafzadegan N. Barthel Index in a Middle-East Country: Translation, Validity and Reliability. Cerebrovasc Dis 2006; 22(5-6):350-4.

13. Hatano S. Experience from a Multicentre Stroke Register: a Preliminary Report. Bull World Health Organ 1976; 54(5):541-53.
14. Foroughan M, Jafari Z, GhemMagham Z. Normalization of IQCODE and AMTS Tests among Aged People and Correlation of Their Results with MMSE Test, University of Social Welfare and Rehabilitation Sciences. Research Grant. 2006. [Text in Persian]
15. Jorm AF. The Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE): a Review. Int Psychogeriatr 2004; 16(3):275-93.
16. Tang WK, Chan SS, Chiu HF, Wong KS, Kwok TC, Mok V, Ungvari GS. Can IQCODE Detect Poststroke Dementia? Int J Geriatr Psychiatry 2003; 18(8):706-10.
17. Barba R, Martinez-Espinosa S, Rodriguez-Garcia E, Pondal M, Vivancos J, Del Ser T. Poststroke Dementia: Clinical Features and Risk Factors. Stroke 2000; 31:1494-501.
18. Lemeshow S, Hosmer DW, Klar J, Lwanga SK. Adequacy of Sample Size in Health Studies. WHO 1990:1-4.
19. Kasner SE, Chalela JA, Luciano JM, Cucchiara BL, Raps EC, McGarvey ML, et al. Reliability and Validity of Estimating the NIH Stroke Scale Score from Medical Records. Stroke 1999; 30(8):1534-7.
20. Mahoney FL, Barthel DW. Functional Evaluation: The Barthel Index. Md State Med J 1965; 114:61.
21. O'Sullivan SB, Schmitz TJ. Physical Rehabilitation. 5th. Philadelphia: F.A: Davis Company; 2007.
22. Delavar A. Applied Probabilities and Statistics in Psychology and Educational sciences. 8th Ed. Tehran: Roshd Publishers; 2005.
23. Streiner DL, Norman GR. Health Measurement Scales: A Practical Guide to Their Development and Use .New York: Oxford University; 2003.
24. Salter K, Jutai JW, Teasell R, Foley NC, Bitensky J, Bayley M. Issues for Selection of Outcome Measures in Stroke Rehabilitation: ICF Participation. Disabil Rehabil 2005; 27(9):507-28.

25. Frankel MR1, Morgenstern LB, Kwiatkowski T, Lu M, Tilley BC, Broderick JP, et al. Predicting Prognosis after Stroke: a Placebo Group Analysis from the National Institute of Neurological Disorders and Stroke rt-PA Stroke Trial. Neurology 2000; 55(7):952-9.
26. Meyer BC1, Hemmen TM, Jackson CM, Lyden PD. Modified National Institutes of Health Stroke Scale for Use in Stroke Clinical Trials: Prospective Reliability and Validity. Stroke 2002; 33(5):1261-6.
27. Lyden P, Lu M, Jackson C, Marler J, Kothari R, Brott T, et al. Underlying Structure of the National Institutes of Health Stroke Scale: Results of a Factor Analysis. NINDS tPA Stroke Trial Investigators. Stroke 1999; 30(11):2347-54.

28. Lyden P, Claesson L, Havstad S, Ashwood T, Lu M. Factor Analysis of the National Institutes of Health Stroke Scale in Patients with Large Strokes. Arch Neurol 2004; 61(11):1677-80.

29. Saver JL, Johnston KC, Homer D, Wityk R, Koroshetz W, Truskowski LL, et al. Infarct Volume as a Surrogate or Auxiliary Outcome Measure in Ischemic Stroke Clinical Trials. The RANTTAS Investigators. Stroke 1999; 30(2):293-8.