Progress in science is dependent upon, and frequently follows, the development of new measurement techniques. In the context of controlled trials of physiotherapeutic techniques, the major requirements are that any measure should be: valid, reliable when used by different observers, simple enough to be used on patients who are often old and suffering other problems, and sensitive enough to detect clinically significant differences. This paper discusses measures of arm function which might fulfil these criteria.

SUMMARY Seven clinical tests have been used to study the recovery of arm function in 92 patients over 2 years following their stroke. These tests are simple and quick, and can be used by any interested observer. They form a hierarchical scale that measures recovery. Statistically significant improvement is only seen in the first 3 months. Fifty-six patients initially had nonfunctional arms; eight made a "complete recovery" and 14 a partial recovery. The tests described are inadequate on their own because they are not sufficiently sensitive at the upper range of ability. While recovery of lost function does relate to the degree of initial neurological loss in the arm, it seems to be largely independent of the overall severity of the stroke.Recovery after stroke dan be measured in many ways, and the method chosen will depend upon the information wanted. Survival apart, the simplest ways can include the length of stay in hospital and final "placement" (type of accommodation), but these depend upon social factors as much as upon the degree of physical recovery. "Activities of Daily Living" (ADL) scales (for example the Barthel') relate much more to the patient alone, and have practical and prognostic value. However, using these scales it is difficult to separate the recovery of specific lost neurological function from a more generalised adaptive response.The study of isolated arm function might allow measurement of the recovery of lost neurological ability separate from the adaptive response adopted using preserved functions (for example, learning to eat and dress one-handed). If so, one could assess the effect of therapeutic intervention (for example, physiotherapy) upon recovery itself. In addition, it could help provide a prognosis for recovery.Arm function after stroke has been measured previously,2-' but no single technique has become generally used. Previous methods have either depended upon special equipment,2 or required time consuming assessments3-5 to be made. Therefore these techniques cannot easily be used in large scale follow-up studies. There is a need for a simple method of measuring arm function that can be used upon a wide range of patients by any interested observer using the minimum of equipment. This paper reports upon a method that has been used in a long-term follow-up study on all patients referred to the Frenchay Stroke Unit over the 2 years 1976-78. PatientsOne hundred and sixty two patients were accepted by the Unit, a specialist rehabilitation department; all referrals were accepted provided that the patient lived in or near the district, and were fit enough to attend. The total includes not only patients admitted to the hospital but also 44 acute strokes who were never admitted to hospital but received all their treatment as out-patients.Ninety two of the 101 patients who survived to their final follow-up were assessed, and they form the basis for the results, except where stated otherwise. There were 45 men and 47 women; 49 had a right hemiplegia, 41 a left hemiplegia, and two had "brai...

This paper describes preliminary studies on a screening test for aphasia which takes 3-10 minutes to complete and which is suitable for use by general practitioners, junior medical staff and other non-specialists. Data are presented to show that it is a reliable, valid assessment. Using cut-off values derived from normal people, the test is sensitive, but its specificity is limited by such associated factors as hemianopia. Using cut-off values derived from patients known to have aphasia, its specificity is improved. An abnormal result needs to be interpreted in the light of all available clinical information. The test should help identify patients with linguistic disturbance.International Rehabilitation Medicine 1986.8:166-170.

The community scheme for stroke survivors was a low-cost intervention successful in improving physical integration, maintained at one year, when compared with standard care.

SUMMARY Ninety-nine patients had their function recorded regularly over the first 13 weeks after their stroke. Five functional areas were studied: urinary continence, mobility, the ability to dress, feeding, and the ability to transfer from bed to chair. Thirty-two patients died before 13 weeks. Forty-five of the 67 survivors had assessments twice weekly from within 4 days of their stroke. Recovery in these 45 patients occurred fastest in the first 2 weeks, by which time at least 50% of recovery had occurred, but it was still continuing at 13 weeks. Urinary incontinence present between 7 and 10 days after stroke was the most important adverse prognostic factor both for survival and for recovery of function. Age was the second most important factor. Hospital discharge seemed to occur once recovery had stopped, although four of the 49 patients discharged had been fully independent for at least 12 days prior to discharge. It is suggested that rehabilitative therapy should concentrate less on physical function and more on cognitive ability.