The authors concluded that evidence on the effects of tilted seat positions on functional and social outcomes among non-ambulant people with neurological or neuromuscular impairment is lacking. A posterior tilt may reduce pressure under the pelvis in those with neurological impairment. The reliability of these conclusions is unclear given the poor quality and variability of the primary studies.

To evaluate the effect of a tilted seat position for non-ambulant adults and children with neurological or neuromuscular impairment.

MEDLINE (1950 to 2006), EMBASE (1980 to 2006), CINAHL (1982 to 2006) and AMED (1985 to 2006); the search terms were reported. The references of retrieved studies and reviews were also examined, and experts in the field were consulted for unpublished studies. The search was limited to articles in English.

Eligible studies investigated the effect of seat tilt, compared either with a fixed seat position or with different angles of tilt. Tilt was defined as rotation of the whole seat around a mediolateral axis. The included studies utilised wheelchairs, chairs and an adjustable bench. These tilted either anteriorly and/or posteriorly by up to 30^ο or 45^ο (respectively). Some had supports for the head, feet, trunk or pelvis, and/or arm rests, cushions and wedges. The duration of the interventions varied from a few seconds to 20 minutes, sometimes with an acclimatisation period. Some studies included additional interventions such as extra seat positions, different wheelchairs or different cushions: these interventions were not reported in the review. The participants in eligible studies were non-ambulant adults or children with an acquired or congenital neurological or neuromuscular condition. All participants in the included studies had neurological impairment. The studies included children aged 2 to 18 years (where stated) with cerebral palsy, neural tube defect or unspecified neurological impairment, and adults with spinal cord injuries or multiple sclerosis. Studies that reported the effect of seat tilt on body physiology, body parts or systems, and human activity (e.g. social participation), were eligible. The outcomes in the review included interface pressure, posture and stability, muscle activity, respiratory function and other functional activities. No outcomes relating to human social activity were reported. Randomised and non-randomised studies were eligible for inclusion.

Two reviewers independently selected studies for inclusion.

Study validity was assessed according to the following criteria: use of controls, randomisation, allocation concealment and equality of groups at baseline. Two reviewers independently carried out the assessment.

Continuous data were extracted as mean percentages of change from baseline with associated p-values and/or mean differences from baseline with associated standard deviations. Where a standard deviation was not reported, a conservative calculation was made, based on data from another study. Where studies conducted multiple interventions among the same participants, best- and worst-case scenarios were calculated.

Two reviewers independently extracted the study data. Study authors were contacted for further information, where required.

The results were combined in a narrative, grouped by outcome. For one outcome, relevant studies were pooled in a random-effects meta-analysis to calculate a weighted mean difference (WMD) with 95% confidence intervals (CIs). Clinical and methodological heterogeneity between the studies was considered in the text.

Twenty studies were included, two of which were reported in the same publication. There were 9 quasi-experimental crossover studies (n=130), 8 uncontrolled observational studies, all using a crossover design (n=103), and 3 studies graded as expert opinion (one case series, n=23; one case study, n=1; one crossover study, n=2). Eleven studies were described as using randomisation.

The studies were heterogeneous with regard to their design, populations, interventions, methodology and outcomes. None of the comparative studies had more than 20 participants, none were blinded, and 9 studies were not randomised. As most of the non-randomised studies delivered interventions in a set order of tilt orientations, fatigue was a potential confounding factor. Study protocols failed to control for other influences on posture.

Interface pressure.

Six studies reported this outcome, of which 5 crossover studies (n=71) were suitable for pooling; two of these were randomised (n=33). Populations and tilt angles differed. When participants were posterior tilted 20^ο to 45^ο, there was a statistically significant reduction in pressure under the ischial tuberosities, compared with the use of the upright position (worst-case scenario WMD 24.00 mmHg, 95% CI: 4.19, 43.80, p=0.02; best-case scenario WMD 24.80 mmHg, 95% CI: 7.16, 42.44, p=0.006).

Muscle activity (5 studies, n=77).

Three studies (n=52) conducted among participants with spinal cord injuries and cerebral palsy reported that tilting the seat affected electromyographic activity. It was unclear whether this had a positive or negative overall effect on functional activities. The participants' ages and conditions varied widely, as did tilt angles and directions.

Other outcomes.

Results for postural and stability measures (6 studies, n=84 to n=104), respiratory function (2 studies n=16 and n=36) and other functional activities (6 studies, n=74 to n=94) were inconsistent. Population ages and conditions varied widely, as did tilt angles and directions.

Other results were reported in the review.

Evidence on the effects of a tilted seat position on health, functional and participation outcomes among non-ambulant adults and children with neurological or neuromuscular impairment is lacking. A posterior seat tilt may reduce pressure under the pelvis in populations with spinal cord injury and neural tube defect.

The review question was clear and the search was adequate in most respects, although the restriction to studies in English might have meant that some studies were missed. The inclusion criteria were wide, especially with regard to the participants and study design. Steps were taken to minimise error and bias in the review process, such as having more than one reviewer independently select studies, assess validity and extract the data; however, it was unclear how any discrepancies were resolved. The gradings allocated to primary studies in the summary table were inadequately explained and do not appear congruent with the table of levels of evidence. The statistical methods used for meta-analysis were designed for parallel data and appear unsuitable for crossover studies as they involve double-counting. Where the results were reported in the summary table, it was not always clear whether they were statistically significant and, if so, which group was favoured. Methodological and clinical heterogeneity were addressed in the text, but there was no indication that statistical heterogeneity or publication bias were assessed. The primary studies were small and of limited quality. The wide variation among populations, interventions and outcomes included in the review make the results difficult to interpret. The reliability of the authors' conclusions is unclear given the heterogeneity and poor quality of many of the included studies.

Practice: The authors stated that a posterior seat tilt of 20^o or more may reduce pressure under the pelvis in populations with spinal cord injury and neural tube defect.

Research: The authors stated that the effects of a tilted seat on populations with progressive neuromuscular conditions warrants study, as no evidence was identified on this population. They recommended use of the Gross Motor Function Classification System for future studies of cerebral palsy in children. They also recommended that further research on tilted seat positions should involve long-term sitting and should include health, functional and social outcomes (including pressure ulcer prevalence, fatigue and ease of transfer using a hoist).

Not stated.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.