5. Results of observational studies.

Demir 2008	Statistical analysis	Wilcoxon matched‐pairs signed‐rank test, Mann‐Whitney U test, Spearman analysis.
	Results	Primary outcome: not addressed. Secondary outcome: Activity limitation ‐ There were improvements in reduction of disability as measured by FIM gains (mean ± SD) from admission to discharge (33.2 ± 12.7,  p = 0.001) and also at 6 months after rehabilitation (20.9 ± 13.4, P = 0.001). GBS patients had poorer health‐related quality of life at six months after rehabilitation compared with healthy controls. The scores of all of the NHP domains (intervention mean ± SD vs control) were statistically significantly higher in the patients (physical mobility 51.8 ± 29.7 vs 5.3 ± 5.6, energy 40.7 ± 35.8 vs 9.2 ± 11.3, pain 20.5 ± 20.2 vs 5.2 ± 11.4, sleep 38.9 ± 32.8 vs 11.7 ± 20.3, social isolation 34.9 ± 32.8 vs 5.2 ± 9.7 and emotional reactions 31.5 ± 23.7 vs 9.5 ± 21.4, P < 0.001 for all domains). Functional scores both at discharge and at the 6‐month follow‐up were highly related to the NHP scores (P < 0.05 across all domains). In particular, the correlations between FIM scores and the energy level (r = ‐0.58 at discharge), physical mobility (r = ‐0.61 at discharge and r = ‐0.48 at 6 months) and emotional reaction (r = ‐0.41 at 6 months) domains of NHP were highly significant (P < 0.01). Being a female, employment, mechanical ventilation, a tendency to depression and educational status were found to be associated with several NHP domains (P < 0.001). Age and marital status showed no association with the NHP scores.
	Author’s conclusions	The HRQOL (health‐related quality of life) of GBS patients remains lower than that of the healthy control subjects. There was a significant improvement in the functional scores in GBS patients at discharge and 6 months after rehabilitation. The reduced HRQOL after GBS appears to be related not only to the physical disability, but also to several demographic and medical variables, such as educational level, employment, gender, mechanical ventilation and psychological factors.
Meythaler 1997	Statistical analysis	Chi2 test, one‐way ANOVA, Pearson’s correlation coefficients.
	Results	Primary outcome: not addressed. Secondary outcome: Activity limitation < 12 months ‐ There was reduction in disability as FIM scores improved from admission to discharge: FIM motor score improved by an average of 15.7 points from mean admission score of 34.7 to mean discharge score of 50.3. The FIM cognitive score improved by an average of 6.9 points from mean admission score of 78.1 to mean discharge score of 85.
	Notes	Other outcome measures that were reported in this study included length of acute hospitalisation, length of inpatient rehabilitation, and acute and rehabilitation charges (costs). These outcome measures have not been included in this table since they were not pre‐selected measures for the review.
	Author's conclusions	The requirement of prior ventilator support most strongly predicts an extended length of stay for inpatient rehabilitation and had most significant impact on total hospital length of stay (both acute and rehabilitation) and the amount of the hospital charges.
Nicholas 2000	Statistical analysis	Multiple regression analysis
	Results	Primary outcome: not addressed. Secondary outcome: Activity limitation < 12 months ‐ There was improvement in BI and FIM scores at discharge compared to admission scores: Mean modified BI score increased from 10 to 19; Mean FIM score increased from 53 to 85. Quality of life < 12 months ‐ Environmental Status Scale (ESS) (n = 15) decreased from 23 to 13, Handicap Assessment Scale (HAS) (n = 9) decreased from 19 to 10 indicating a reduction in handicap.
	Author’s conclusions	Significant improvement in function and reduction in handicap occurred during rehabilitation, which was demonstrated by standardised outcome measures.