Introduction

Spinal cord injury (SCI) affects many individuals across the world. The annual incidence of traumatic SCI is estimated to range between 12.1 and 57.8 per million worldwide,1 and SCI rates vary between countries and across regions—in North America (39 per million), Western Europe (16 per million), Australia (15 per million) and in New Zealand (30–49 per million).2, 3 In Australia, this equates to around 285 new acute traumatic SCIs per year, predominantly from transport-related accidents (46%) and falls (28%), with tetraplegia (53%) slightly more frequent than paraplegia and the majority of cases involving males, with a male to female injury ratio of 5.3:1.4 Non-traumatic SCI adds to these figures; an Australian study reported an age-adjusted adult incidence rate of 26.3 cases per million per year.5 SCI prevalence is estimated to be 10–12000 traumatic and 8000 non-traumatic cases in Australia.6, 7

Sensorimotor and autonomic nervous system dysfunction following SCI results in a range of acute, rehabilitation and long-term healthcare challenges, including pressure injuries, disorders of muscle tone and bowel and bladder problems.8 These sequelae have long-term effects on independence and psychological well-being post SCI.9 Aside from the potentially devastating impact of traumatic SCI on physical function, social participation and quality of life, traumatic SCI carries a high financial cost, estimated in Australia to be $2 billion annually, or $5 million per case of paraplegia and $9.5 million per case of tetraplegia.6

Research prioritisation has become an area of interest in recent years, owing to a high demand for evidence-based resources combined with limited research resources.10, 11 Research prioritisation is of particular importance in SCI, given that the breadth of challenges experienced by people after SCI presents multiple potential avenues of enquiry. The process of prioritising research is complex. There are numerous prioritisation criteria including clinical importance/magnitude of problem, likelihood of reducing burden, cost-effectiveness, present knowledge, resources, ethical aspects, research capacity, novelty and controversy.12, 13 These can be broadly categorised into three key domains: public health benefit (should we do it?); feasibility (can we do it?); and cost.12 There are also multiple methods for deciding on priorities as outlined by Viergever et al.,12 which can be classified broadly into consensus-based approaches (driven by stakeholder input), metrics-based approaches (for example, the Delphi technique) and combination approaches (for example, prioritisation followed by discussion using a nominal group approach). Two recent examples of SCI research prioritisation are Guest et al.’s14 description of the prioritisation process followed by the North American Clinical Trials Network, which is based on earlier work by Kwon et al.15, 16 who developed a scoring system for grading pre-clinical literature on neuroprotective treatments for acute SCI.

The purpose of this review was to gain an overview of the volume, nature and findings of studies that aimed to identify priorities for SCI research. This was undertaken as part of a structured process of evidence synthesis (rapid review) and stakeholder consultation (expert opinion) to develop a regional (Australia and New Zealand) SCI research strategy. Two companion papers describe subsequent steps in the process.60, 61

Materials and methods

The review used a rapid review methodology. Rapid reviews are an emerging method of efficiently synthesising research evidence in settings such as health policymaking, in which a broad overview of evidence is required in a short time frame, for example, 5 weeks, rather than the 6–24 months required for a systematic review (SR). Rapid reviews are primarily distinguished from SRs by their focus on searching for and summarising synthesised research evidence (that is, relevant reviews) and, where these are not available, high-quality or recent primary studies.17

Search strategy

A comprehensive search of the following databases (from initiation until 31 October 2012) was undertaken: Medline (1950–31 October 2012; see Appendix 1); All EBM (All EBM Reviews: the Cochrane Database of Systematic Reviews, ACP Journal Club, Database of Abstracts of Reviews of Effects (Cochrane methodology register, Health technology assessment, NHS economic evaluation database) and the Cochrane Central Register of Controlled Trials.); CINAHL; PsycINFO; EMBASE; and Web of Science. Google Scholar was also searched using the terms ‘spinal cord injury’ combined with ‘research priorities’ with no date restriction. The first 100 results from the Google Scholar search were screened. Reference lists of included studies were also scanned to identify further relevant references.

Inclusion criteria

Identified citations and full text studies were screened against the following inclusion and exclusion criteria: this was conducted by two authors (LP and PB). An initial 10% of the screening was performed by both authors independently in order to refine interpretation of the inclusion and exclusion criteria.

Inclusion criteria:

  • Population: key stakeholders in SCI research: patients, patient representatives, families and carers; researchers (all types of research); clinicians (all phases of SCI care); policymakers; research funders; and representatives of healthcare organisations.

  • Intervention: any project in which the primary aim was to identify research priorities, or priorities that can be addressed by research, through direct engagement with the above population. Project methods could be qualitative or quantitative (for example, survey).

  • Study type: SRs, defined as ‘…an overview of primary studies which contains an explicit statement of objectives, materials and methods and has been conducted according to explicit and reproducible methodology.’18 p. 672.

  • Publication status: published in peer-reviewed journals.

Exclusion criteria

  • Primary studies or reviews in which there is no direct engagement with the population of interest. SRs often contain discussion regarding future research priorities, however, these are influenced by review scope and reflect primarily the views of the review authors.

  • Studies examining research priorities within a specified area of SCI (for example, pressure ulcers, bladder care), because the review focus was the entire field of SCI.

Quality appraisal

Quantitative studies collect and analyse numerical or categorical data, whereas qualitative research emphasises in-depth exploration and description rather than numerical measurement.19 Therefore, different methods are used to both review and evaluate reviews of quantitative and qualitative literature. Eligible quantitative SRs were critically appraised using the AMSTAR tool (http://amstar.ca/), an 11-item tool with well-established validity and reliability that is extensively used to evaluate quantitative SRs.20, 21, 22 Critical appraisal of qualitative studies is an area of ongoing debate, and currently there is no consensus on an appropriate critical appraisal tool for qualitative research, and researchers are advised to choose a tool specific to this research paradigm.23 Therefore, eligible qualitative SRs were critically appraised using five criteria designed to evaluate the rigor of qualitative reviews.24 The 11 AMSTAR items and the five qualitative review evaluation criteria, as well as the results of quality appraisal, are contained in Appendix 2. All critical appraisal was carried out by one of the authors (LP).

Results

The search of electronic databases yielded 293 articles. Two reviews met the selection criteria: one SR of quantitative studies25 and one meta-synthesis of qualitative studies.26 The SR of quantitative studies25 was rated as being of low-to-moderate methodological quality; a key shortcoming of this review was lack of quality appraisal of the included studies, which may bias review findings and conclusions. The meta-synthesis of qualitative studies26 was rated as being of high methodological quality, fulfilling all quality assessment criteria (Appendix 2).

Table 1 presents a summary of review characteristics. Table 2 lists the primary review findings, highlighting areas of overlap and the authors’ conclusions. Table 2 shows that all of the priorities identified by Simpson et al.25 were encompassed within the themes identified by Hammell.26 Specifically, the four ‘function’ themes in Simpson et al. relate to concepts 1 and 2 in Hammell; themes 5 and 6 from Simpson et al. relate to concepts 3 and 4 in Hammell; and the remaining concepts in Hammell are not directly relatable to any of the themes from Simpson et al. Characteristics of the studies included in the two reviews (citation, number of participants, data collection method and population) are contained in Appendix 3. Of the 31 included studies within the two reviews, only one27 study involved participants other than people with an SCI. A further two relevant primary studies were identified that had not been included in the reviews,28, 29 which also engaged solely with people with a SCI.

Table 1 Summary of included systematic reviews
Table 2 Findings of eligible systematic reviews illustrating areas of convergence

Discussion

This is the first known overview of reviews examining research priorities in the field of SCI. The rapid review approach is less comprehensive and robust than a SR. This means that caution needs to be applied when interpreting review findings, as deeper exploration of primary literature may elucidate further insights and therefore influence interpretation of review findings. However, rapid reviews, because of their timeliness, have a number of potential uses:

  • ‘to serve as an informative brief that prepares stakeholders for discussion on a policy issue;

  • to support the direction and evidence-base for various health policy initiatives; and

  • to support the development of clinical interventions and/or health services programs.’ (Khangura et al. p. 2).17

These uses aligned well with the overarching purpose of this project. Specifically, the findings of this rapid review were used to inform subsequent steps in the development of a regional SCI research strategy, as described in two companion papers.60, 61

Synthesis of the findings from these reviews is challenging, as they were conducted on separate bodies of literature, using different research paradigms. The reviews had to be evaluated with separate quality appraisal tools for quantitative and qualitative research synthesis (Appendix 2), meaning that a direct comparison of the results of quality appraisal is not possible. Notwithstanding the important differences in study methodology between the included reviews, there was some convergence in review findings. Simpson et al.25 noted that Hammell’s qualitative review26 identified ‘6/10 themes directly related to physical, social and psychological areas’ (p. 1554). These are not specified, but two themes from Hammell’s review appear directly comparable to the priorities highlighted by Simpson: ‘problems associated with an impaired body,’ and ‘renewed importance of relationships.’ This convergence of review findings is interesting in the context that the reviews were published 5 years apart. This could suggest that priorities identified in Hammell’s review26 have not been fully addressed by subsequent research, and it raises the question of how identified priorities can be (or are being) fed into the research process.

Another way to interpret the results of these reviews is to view them as contributing complementary perspectives. Quantitative studies offer greater explanatory power as they generally involve larger cohorts, whereas qualitative studies provide a more in-depth examination of issues through exploratory interviews on smaller cohorts. Simpson et al.’s review of 24 quantitative studies with a combined sample of over 5000 participants highlights four areas of function—bowel, bladder, sexual and motor—and two life domains—health and relationships. The authors of this review conclude that this information can inform research planning by helping to align consumer and researcher priorities.25 Hammell’s26 qualitative review of 7 studies with a combined sample of 120 participants and a focus on exploration of factors that influence Quality of Life offers ‘a more nuanced understanding of the experience of Quality of Life SCI than is achievable by quantitative methods’ (p. 136). Specifically, qualitative exploration facilitates description of important concepts associated with adjustment following SCI such as self-worth, injury and loss, and development of new values and perspectives. It also enables relationships between identified priorities to be elucidated. Such information can not only inform the choice of research topics but can also be valuable in different ways for research planning—for example, to identify outcome measures that are meaningful to people with a SCI, regardless of research topic. Future primary studies could incorporate mixed qualitative and quantitative methods in order to harness the complementary information gathered by these approaches.

This review found that only 1 of 31 primary studies examining SCI research priorities included participants other than people with an SCI. Harmonising the lived experience of SCI with research, clinical, policy and other inputs ensures the development of robust, feasible and high-quality research that is more likely to ultimately translate to clinical practice and health policy change. For example, pressure injuries are not among the highest priorities identified by people with SCI. However, severe pressure injuries carry a high burden for both people with SCI and the wider healthcare system. Therefore, pressure injuries may be considered a higher research priority from a clinical and health economics perspective. This illustrates the importance of identifying and balancing multiple perspectives when framing a strategic research agenda.

In addition to gathering the perspectives of people with SCI on research priorities, future studies should also involve other relevant SCI research stakeholders including carers, clinicians, researchers, policymakers and research funders. To this end, we brought together a representative group of these stakeholders to determine an overarching SCI research strategy that addresses how best to prioritise, plan and successfully undertake SCI research so that its benefits can be fully realised for those living with SCI. This work is described in two companion papers to this review—one outlining the background materials and methods 60 and one presenting the results of a day-long stakeholder dialogue that deliberated upon these issues.61