Population-Based Spinal Cord Injury Registries: Potential Impacts and Challenges

Marcel F. Dvorak, Catherine A. McGuinness, Michael G. Fehlings, and Vanessa K. Noonan

Key Points

1. Prospective data registries provide an effective way of tracking outcomes and epidemiology and can be useful in collecting valuable observational data that can inform clinical practice.

2. A registry should be focused on key data elements to be useful.

3. To ensure that the data collected remain focused, specific research questions should be formulated when setting up the registry.

4. Challenges in setting up such registries can include issues of privacy related to collecting data on patients, which is a further incentive to have focused research questions making the value and intended use of the data clear to those concerned.

As research funding becomes scarcer, the focus and practical impact of clinical research activities are coming under increased scrutiny by the agencies that provide funding for research.1 Governmental funding agencies, industry, and nonprofit foundations are more cognizant of the outputs and outcomes of the research that they do support and are more likely to consider investing in research activities that are mission driven, focused on a clinical need, and strictly managed, as opposed to more exploratory endeavors. In this environment, it is more difficult to obtain financial and conceptual support for population-based observational studies that are often the outputs of patient registries. Patient registries, therefore, must be clear with respect to their outputs and deliverables.

Advances in the care of acute traumatic spinal cord injury (SCI) have meant that patients who experience SCI are now expected to survive and in many cases recover from their injuries.² The Rick Hansen Spinal Cord Injury Registry (RHSCIR) is a pan-Canadian patient registry that was initiated in 2002. This chapter describes why registries are important, what unique contributions the RHSCIR can make to improving patient care, and some of the challenges in establishing the RHSCIR, specifically, the impact of new privacy legislation on the registry, the influence this has had on the quality and quantity of data obtained, and how the RHSCIR has addressed these privacy issues. The RHSCIR will be used as an example of a population-based prospective data collection project; however, we acknowledge that other organizations, such as the North American Clinical Trials Network (NACTN) and the European Multicentre Study about Spinal Cord Injury (EMSCI), are establishing similar datasets.

Why Is It Important to Have a Spinal Cord Injury Registry?

A patient registry is defined as “an organized system that uses observational study methods to collect uniform data (clinical and other) to evaluate specified outcomes for a population defined by a particular disease, condition, or exposure, that serves predetermined scientific, clinical, or policy purpose(s)” and a “patient registry database describes a file (or files) derived from the registry.”2 The work of a registry is to identify and encourage the use of therapies and techniques that are effective, while minimizing reliance on those that are not. In the environment of cost containment and rapidly advancing technological capability, there is a strong need to determine which acute care, rehabilitation, and postdischarge community therapeutic interventions will lead to improved long-term patient outcomes, and to collect observational population-based data with long-term follow-up to allow this to be determined. Costs of acute surgical care and inpatient rehabilitation care, combined with restrictions on access to care, make this a critically important issue.

The collection and reporting of observational data, when combined with advances in medical and technological surgical care over the past half-century, have led to dramatic improvements in the care of patients with acute SCI.³ The care improvements that affect patients include the advent of specialized centers of SCI care^4,5; new techniques for anesthesia,⁶ hemodynamic monitoring,⁷ and improved thromboprophylaxis^8,9; advances in surgical instrumentation; and technical advances that enable safe, early, and definitive surgical decompression and stabilization.^10,11

Registries facilitate the collection of data for long-term observational studies, to determine evidence-based best practices for the treatment and care of those with SCI. Although randomized, controlled trials (RTCs) are considered the gold standard for evaluating new treatments, the RCT approach of creating an “ideal situation” by standardizing all aspects of the study may not always be necessary, appropriate, possible, nor adequate for evaluating treatments and therapies, particularly for SCI, where the numbers of patients are relatively small compared with other diseases and injury conditions.^12–15 In the clinical setting, it is difficult to establish control groups, particularly when invasive or surgical therapies are being examined. A well-designed observational study can assess interventions and practices commonly used in the clinical setting by controlling for known confounders.^16,17 If a registry collects a population-based sample (or one that represents the complete population at least in some geographic locations), then the relevance and generalizability of the data increase dramatically; the data in the registry then represent the circumstances in a day-to-day clinical situation.

One need only look at the impact of several large clinical trials on SCI to question whether they have had any positive influence on patient care or outcomes.^18–22 Infact, the multiple clinical trials on the use of methylprednisolone have, if anything, caused uncertainty among patients, physicians, and the legal community, while the question of effectiveness remains largely unanswered.^19,23,24 At the same time, large-scale observational studies have made substantial advances in answering questions related to therapeutic effectiveness in relation to thromboprophylaxis,^8,9 surgical timing for SCI,²⁵ and other spinal conditions.^26–28 One might argue that most of the substantial advances in patient care have come as the result of clinical and technological advances that have been reported initially as small case series and have slowly gained acceptance despite the low quality of the literature supporting these changes in practice.

What Unique Contributions to Patient Care Can the RHSCIR Make?

The RHSCIR was initiated with the goal of creating a national registry of individuals who sustain an acute traumatic SCI to enable translational research, to promote excellence in clinical practice, to facilitate the translation of therapeutic interventions for SCI, and to foster enhanced quality of life for people living with SCI. The RHSCIR leads and supports the collection and analysis of observational data for the purpose of conducting large-scale observational studies that facilitate the investigation of the potential links between the processes of care delivery and patient outcome. The RHSCIR model of translational research not only includes “the process of applying ideas, insights and discoveries generated through basic scientific inquiry to the treatment or prevention of disease or injury” but also takes into account the flow of information in the other direction, from the patient to inform the researchers. The registry populates economic and business case models, thus adding a dimension of transparent accountability to the researchers and influencing health care policy.

The integration of patient data from one phase of care to the next—prehospital to acute to rehab and then into community—allows health care providers to obtain information on the long-term outcomes of persons with SCI. Often it is the interventions applied or withheld in the first moments following injury (i.e., appropriate immobilization, timely imaging and decompression, etc.) that have a profound influence on eventual outcome. Reporting on a subacute rehabilitation intervention without controlling for and describing the acute interventions that might influence eventual outcome leaves one open to significant bias. Understanding how the patient was affected and influenced by various treatments encourages the validation of existing best practices, and will encourage surgeons and rehabilitation specialists to make best practice treatment decisions for other patients. Furthermore, it will be possible to evaluate the degree to which clinicians are following evidence-based guidelines, and this will provide a mechanism to improve the quality of care provided.

The existence of a national registry ensures that all data collected have standardized definitions and the patient registry database is linked to the primary data source, thereby streamlining data collection, eliminating duplicate data collection, and ensuring a consistent dataset is collected on all patients, which includes information from their first encounter with the medical system, through rehabilitation and community integration, through to periodic updates. From a health services provision perspective, this continuity of data flow is important.

Observational studies and RCTs have complementary roles in evaluating evidence for patient care and treatment, and a registry can enable both types of studies. The development of the RHSCIR has been a long-term project requiring considerable effort to create a national infrastructure. We are currently using the RHSCIR to conduct observational studies and are set to begin to use the RHSCIR program and technology infrastructure to manage the data collection for several clinical trials.

A national registry ensures the quality and consistency of research data. At present, research-minded health care professionals, lacking a centrally organized place to store data, often track treatment in stand-alone, personal databases. Many of these databases are not linked, so that the patient information is not continuous from one phase of treatment to another, patients may be entered multiple times in different datasets, patients are lost when they enter another phase of care or when they move to another community, and the data are collected in a nonstandardized format that makes comparisons between sites impossible. Furthermore, these individual physician- or site-based databases also use unique outcome tools that may not have been validated from a psychometric point of view. The use of nonstandard and unvalidated outcomes further complicates analysis between sites. A common terminology and standardized data collection processes, the use of unique patient identifiers to track patients from one geographical location or phase of treatment to the next, detailed procedures to ensure data quality, and the use of sophisticated technology to create a centralized place for data storage are all critical components of the RHSCIR. We have found that implementing standardized data collection protocols at certain sites, through a process of training and site visits and by mandating the use of standardized motor score assessment and recording, has resulted in substantial changes to the assessment and documentation of spinal cord impairment at these sites. Simply establishing a process for measuring and recording a neurological impairment can result in substantial improvements in the techniques used for neurological assessment at a given site.

The final benefit of a registry is the value that it can offer to patients. A registry allows patients, if they consent ahead of time, to be contacted for continuing care, innovative therapies, and the like. If a new therapy or treatment is developed, eligible patients can be rapidly identified and contacted with respect to the new study or treatment. The RHSCIR Web site will eventually provide information about new technologies and new service initiatives to interested patients. Because the registry acts as a single point of contact between the patients and researchers, the burden to the patient and health care provider is reduced. This single point of contact will aid the patient by answering questions or addressing concerns about the registry and can help to manage appropriate referrals to support services. The registry also informs and educates researchers/clinicians so that more data are collected as part of clinical practice instead of being obtained through separate interviews with patients; this also improves data consistency and promotes evidence-based practice. Finally, the RHSCIR, part of the Rick Hansen Institute (http://www.rickhanseninstitute.org), connects patients to a national network of people committed to reducing the incidence of permanent paralysis, reducing secondary complications related to SCI, and improving outcomes and quality of life. These benefits are listed in Table 51.1.

Functional Aspects of RHSCIR

The RHSCIR is a pan-Canadian initiative, collecting data through a network of 13 local member sites located in eight provinces to date (Fig. 51.1 and Table 51.2). The data collected by the RHSCIR include detailed information on incident cases of traumatic SCI, American Spinal Injury Association (ASIA) Impairment Scale A, B, C, D, and cauda equina injuries, and are aligned with the International SCI Data Sets.²⁹ The data include sociodemographic data, medical history, diagnoses, interventions, procedures, and complications. Data are collected during the acute and rehabilitation phases through to community integration using patient-reported outcome measures at 1 and 2 years postinjury and at 5-year intervals until death. It is expected that the inclusion criteria will extend in the future to those with non-traumatic SCI, as the potential long-term impacts of the registry will benefit all persons with spinal cord impairment and will enroll individuals living with longstanding SCI.

A key challenge of acute traumatic SCI is obtaining consent from patients. Patients with minimal impairment are often discharged before engaging in the informed consent process required by the Personal Information Protection and Electronic Documents Act (PIPEDA) regarding their participation in a registry or research studies. For this reason, the RHSCIR has developed several strategies, including the collection of data from both responding and nonresponding patients. Patients are presented with consent packages, which contain the RHSCIR privacy policies, so that they have a full understanding of what will be done with the data. A coordinator is present to explain and obtain consents from the responding patients, a process that often requires repeated visits to ensure the patient has an opportunity to ask questions and is fully informed before consent is obtained. Patients who are nonresponders, either because they are discharged or chose not to participate, are included in the registry using an abstracted core dataset only; this contains minimal deidentified data about the type of injury to ensure that the registry is collecting population-based information with no selection bias. Participation in the registry at a core dataset level does not require direct personal contact between registry staff and the patient because the data are either collected as a part of the processes of care or abstracted from the medical record. Participation at this level requires no direct involvement of patients beyond their consent to participate. Participation in the full registry requires direct patient contact and consent to obtain accurate sociodemographic and socioeconomic data and to enable subsequent contact with the patient in the community to complete patient reported outcomes during scheduled follow-ups.

Table 51.1 Benefits to Patients as a Result of Participating in a Registry

• Patients may have the opportunity to learn about new therapies or treatments. A Web site can eventually provide information about new technologies and new service initiatives.

• The burden to the patient and the health care provider is reduced by the registry’s acting as the single point of contact between the patient and researchers.

• This single point of contact will aid the patient by answering questions or addressing concerns about the registry and can help to manage appropriate referrals to support services.

• Researchers/clinicians are informed and educated so that more data collected are part of clinical practice instead of obtained from patients. This also improves data consistency.

• All registry consents and information are provided in one package, so coordination and explanation of information is more streamlined.

• Through other activities associated with the Rick Hansen institute (RHI), patients can be connected to people committed to reducing incidence of traumatic spinal cord injury and improving outcomes and quality of life.