Introduction

Spinal cord injury (SCI) is a multi-system, multi-factorial, life changing event for any individual. The impact of SCI on an individual may be viewed from physiological or functional aspects, or from the opportunity or ability an individual has to undertake desired societal roles.

Measurement of such variables in health care after delivery of services is now commonplace within rehabilitation and in health care generally, however it appears rare that such outcomes across all of the aforementioned perspectives are measured. A failure to do this with a multifaceted injury such as SCI, will provide an incomplete picture for those tasked to review and plan services and processes to achieve optimum outcomes.

There are 12 specialist spinal injury units (SIU) in the UK and Ireland, the aim of which, is to provide a comprehensive service which manages the impact of SCI immediately following acute injury, through rehabilitation to a meaningful discharge back into the community. It is clear however that some people with SCI do not have the opportunity to access this system but are managed in a non-specialist environment, commonly orthopaedic or general rehabilitation areas. There has been anecdotal evidence for some time suggesting that outcomes for those managed in a non-specialist system do not match those who utilise the spinal injury unit systems. No study which measures the efficacy of specialist versus non-specialist management of SCI has previously been performed in the UK. The primary question investigated within this study was to ascertain whether a positive difference in outcome from physiological, functional and social perspectives for the individual with SCI actually exists in those managed through a specialist spinal unit compared to those who were managed through non-specialist systems.

Method

The self selecting sample were recruited through Spinal Injuries Association (UK) (SIA) newsletter ‘Forward’. A self-reporting questionnaire was utilised, with considerations made regarding language used ie terminology familiar to those with SCI. A grid format was predominantly used for responses, to accommodate for potential limited hand function of respondents, although there was the necessity for a limited number of free text sections. The questionnaire was piloted (12 people) to ensure appropriate language and clarity of questions posed, and distributed with the SIA newsletter. Demographic information was requested regarding age, gender, injury cause, level and severity, location of rehabilitation and length of hospitalisation. The measuring tool aimed to measure outcome from health, functional and social perspectives. Comparison of all these outcomes between SIU and non-SIU cohorts was to be made.

Health

Eighteen common secondary complications occurring following spinal cord injury were identified from the literature1,2,3 and the author's previous experience of SCI having worked in the speciality for the previous 10 years, clinically within two specialist units and in related academic positions (Table 1). The sample were asked to indicate if they had experienced each within the last 2 years.

Table 1 Percentage incidence of health complications in tetraplegics and paraplegics for SIU and non -SIU groups
Full size table

Functional outcome

Three outcomes were to be investigated: (a) In what activities of daily life did the participants require assistance, and how much? (b) How much paid/unpaid assistance did participants actually receive? (c) Were participants in control of, and satisfied with, the assistance required?

When scaling the level of assistance required, the Functional Independence Measure (FIM) was reviewed. The FIM4 is designed for recording by health care professionals and requires a formal training programme prior to implementation in an attempt to maximise reliability of responses. As the project required participants to self-report, the number of categories was reduced to an ordinal scale with only four potential responses ie 0, <50%, >50%, 100% assistance required. Although some items were based on FIM, to facilitate self-reporting by participants and comprehensiveness, the language was changed and other items altered to encompass other activities not covered by FIM, making a total of 18 items examined (Table 2).

Table 2 Level of assistance required with various functional activities in SIU and non-SIU groups (all figures expressed as percentage)
Full size table

Social activity

Based on dimensions of societal function ie occupation, mobility and social integration and some elements of the CHART (Craig Handicap Assessment and Recording Technique),5 10 key variables relating to social activity were examined (Table 3).

Table 3 Levels of social activity (expressed as percentages)
Full size table

Life satisfaction

For the purposes of this study two other key elements required additional exploration beyond the information already collected. Firstly an indication of how people felt now and whether that had changed over time utilising a simple four-point fixed choice scale. Secondly the areas of most and least satisfaction were examined within a free-text section, respondents being requested to provide up to five areas for each also stating best and worst. These were post-coded following data collection.

Data analysis was performed using SPSS statistical software package. Statistical significance was set at P<0.05.

Results

Of the sample meeting the inclusion criteria (n=800) 13.6% (n=99) had not utilised an SIU system. Comparison of the sample through statistical analysis using a t-test demonstrated no significant difference in age or date of injury between the SIU and non-SIU cohorts. Utilising a Chi-square test, significant differences in severity (P<0.001) the non-SIU group had less severe injuries, gender (P=0.002) and cause (P<0.001) as the non-SIU cohort had a greater number of females and SCI of non-traumatic origin. In view of this further analysis utilising logistic regression was performed on the data as appropriate to adjust for these other variables.

Health outcome

Chi-square test was performed exploring whether a statistically significant difference existed between the two cohorts. Logistic regression was utilised on those that demonstrated significance. Table 1 illustrates percentage incidence in all the complications explored, and indicates a statistically significant lower incidence in the SIU cohort in five and 10 of the 18 within the tetraplegic and paraplegic groups following analysis. Additionally, there was a statistically significant lower incidence of superficial pressure sore (P=0.048) within the SIU cohort, when taking the sample as a whole.

Results – functional outcome

Within activities explored the SIU cohort required a lower percentage level of assistance. Using a Mann-Whitney U-test, of the 18 activities, at statistically significant lower level of assistance was demonstrated in three activities for the whole tetraplegic group and 15 for the paraplegic group (Table 2). Additionally, when examining those with complete tetraplegic only, a further three activities proved significant ie in grooming (P=0.004), eating (P=0.001), and drinking (P<0.001).

Chi square analysis demonstrated a positive outcome related to satisfaction of the amount of assistance received in the SIU cohort (P=0.017).

Results – social activity

Ten key social indicators were suitable for statistical analysis (Table 3). Again percentage differences suggested better outcomes for the SIU cohort. Statistical significance was demonstrated in three areas specifically within the paraplegic group. Additionally, analysis of the sample as a whole indicated the SIU cohort were more likely to have a partner (P=0.012), be in paid employment (P=0.017) or in voluntary employment (P=0.025). Males with SCI in the SIU cohort were more satisfied with sex, in both the tetraplegic (P=0.006) and paraplegic (P=0.05) groups.

Life satisfaction

There was no statistical significance demonstrated in general life satisfaction between the two cohorts (P=0.07). The life areas that respondents appeared to be most satisfied with were similar and tended to have a social focus eg family, friends, social life. Slight differences were raised in the areas of least satisfaction with 10% more of the non-SIU sample raising ‘health’ as an issue. This seems congruent with the statistically significant differences demonstrated in the health section. Pain was noted by over 12% of the sample as a poor life area. The fact that this was ranked the worst area in the non-SIU group and second worst in the SIU cohort is indicative of chronic pain following SCI remaining a complex and most difficult situation to manage.

Discussion

The evidence demonstrates that health outcomes are improved for those going through a specialist SIU system. The implications for individuals and for increased utilisation of health care services for those not attending an SIU must be substantial. However, the incidence of some complications remains high, and there is no room for complacency within SIU's. Greater collaboration between spinal injury units both national and internationally is needed, not only to facilitate cross centre working and development of services, but also to create the necessary critical mass of people with SCI with which to undertake meaningful research to impact on further improvements in outcome.

From the functional perspective, once again the evidence is highly suggestive that those utilising an SIU system have improved outcomes. The extent to which this was the case was surprising, as the non-SIU cohort had less severe injuries, and therefore one would expect better functional outcomes. A comparison of exact ‘like for like’ may well have provided further statistical significance in favour of the SIU cohort. Again this has personal and economic implications, with the need for more paid assistance in the community for those who are less able.

From the social perspective, the investigator expected little differences to exist between cohorts as the main focus of SIU, as immediate rehabilitation post injury has traditionally had a health and functional focus. Although there appears to be a change to models of rehabilitation delivery which focus more on the reintegration of the individual in his or her community, it may take some time before this impacts more fully on social activity. However, the positive outcomes demonstrated in the SIU cohort provides evidence that utilisation of an SIU system is desirable. Employment levels remain exceptionally poor not only in this population, but for those with physical disabilities generally. The problems created by environmental and attitudinal barriers need to be addressed at social policy level, not only regarding employment but also other access and opportunity issues which limit social opportunity.

Clearly there is much scope for future research across all of the perspectives examined. The author is currently re-examining the same data as a whole sample to investigate where relationships may exist between health, function and social activity eg between health and employment. Reasons for the high level of complications apart from the physiological changes following SCI requires investigation, eg are such individuals educated but choose not to comply with methods to minimise the likelihood of the development of such complications due to potential impact on other life activities? The economic impact requires investigation, as use of resources for health care in this population is surely not optimised in the treatment of such health problems.

As discussed previously the continually disturbing employment statistics for this population require further investigation and radical action.

Conclusion

This study has provided conclusive evidence that from health, functional and social perspectives an individual should be managed in a specialist SIU system. This is despite the fact that the SIU cohort showed a more significant severity of injury. Should a like for like comparison have been made as regards severity (completeness) of injury, the likelihood is that the results would have been even more in favour of the SIU cohort. Within the UK, responsibility for ensuring people are referred to a spinal unit must rest with commissioners of health care. The message must get through to referring hospitals not only through dissemination of this study, but also through directives from health authorities to hospitals in their region. From an international perspective those countries developing systems designed to meet the needs of the spinal cord injured, in the light of this evidence, should ensure specific specialist systems are developed.

figure 1

Figure 1