Clinical research methodology in spinal trauma patients: natural experiments and clinical equipoise

This thesis focused on clinical research in spinal trauma patients concentrating on a novel study methodology alternative to the Randomized Control Trial (RCT) design.
In Chapter 2 we present the results of a national survey amongst surgeons responsible for spinal trauma patients in different trauma centers in the Netherlands. For the future, research should focus on enhancing pre-hospital triage.
In Chapter 3, the optimal conservative treatment of thoracolumbar spine fractures is explored. Randomization created comparable groups except for the mean age of female patients with a burst fracture that was significantly higher than male patients. According to the VAS, 20 (18%) of the 108 patients with compression fractures suffered from moderate or severe back pain at long-term follow-up; 12 patients had an Oswestry Disability Score (ODI) greater than 40 indicating moderate disability. Of the 25 patients with burst fracture, 3 (12%) had chronic moderate pain and one patient was operated on because of severe persistent pain.
In Chapter 4 we explore the existing outcome measurements of spinal trauma patients at that time. We used the WHO’s International Classification of Functioning, Disability and Health (ICF) as an expansive theoretical underpinning for newly developed measures targeting, among others, trauma patients. We proposed fourteen domains of relevance in our opinion: mental health, sensory and pain, genitourinary and reproductive, neuromusculoskeletal, nervous system, movement related structures, mobility, self-care, domestic life, interpersonal relationships, major life areas, community-social-civic life, support and relationships, services systems and policies.
In Chapter 5 and 6 the concepts of equipoise and clinical equipoise are introduced.
Clinical equipoise, as introduced by Freedman in 1987, exists when there is a genuine uncertainty within the expert medical community about the optimal treatment of a certain disease.
This concept was the starting point of a cohort study where patients are retrospectively identified and prospectively followed where clinical equipoise exists about the best treatment for patients with a spinal fracture, operative or non/operative. This clinical equipoise concept was put into practice and 636 patients were identified in two hospitals, where in 190 patients there was a discordant treatment preference by two spine surgeons as representatives of the two hospitals. The two treatment groups both consisted of 95 patients and were essentially comparable. Overall outcome of non-operative and operative treatment in middle-long-term follow up was comparable, although there seems to be a difference in neurologic recovery patterns in favor of operative treatment.
In Chapter 7 and 8 the Natural Experiment concept is introduced as inclusion criterion in prospective and retrospective studies based on clinical equipoise. For future natural experiment studies, in public health but also in comparative medical studies, registries of currently available longitudinal (and ideally hierarchical) data systems could add value for enabling natural experimental studies. Our recent review of the NE methodology in spinal trauma only identified 4 articles that used the NE design in the past 19 years. One article used an expert panel, something we would recommend in using the natural experiment methodology in clinical research. Although 4 articles is a disappointing number, the MINORS criteria of the 4 papers showed that the quality of these reports was high, especially compared to other retrospective comparative study designs.