A new stroke upper limb rehabilitation program was designed. Participants were classified into 3 categories based on their level of upper limb impairment reflected by the baseline S-FM score (severe 0-4, moderate 5-8, mild 9-12). Upper limb training was provided according to the standardized treatment choices preset with expert opinions

Treatments were prescribed based on the corporate stroke rehabilitation protocol of Hong Kong Hospital Authority. The protocol allowed variations in clinical practice and ultimate decision about a particular clinical treatment depended on each individual patient's condition and clinical judgment of individual therapists.

Intervention selection included stretching and mobilization exercise, strengthening exercise, electrical stimulation, acupuncture, functional or task specific training and mirror therapy.

S-FM consisted of six items examining upper extremity movements (shoulder elevation, shoulder flexion 90° to 180°, elbow extension, elbow 90° pronation/supination, elbow 90° wrist flexion/extension and grasp, adduct thumb). Each item was scored with an ordinal scale of 0 (unable to perform), 1 (partially performed) and 2 (fully performed). The total score was 12.

FTHUE consisted of 18 activities examining upper extremity function. Seven levels with 1-3 activities in each level in a hierarchy sequence of difficulty were adopted.

MBI consisted of 11 items examining ADL including self-care, continence and locomotion. A modified scoring system with 5 codes (unable to perform task, attempts task but unsafe, moderate help required, minimal help required and fully independent) was adopted for each item, giving a total score from 0 to 100 (0= fully dependent, 100= fully independent).