Big data in healthcare: are we close to it? - Critical Care Science (CCS)
Translating medical research into clinical practice guidelines is not trivial. There has been a surge in the number of published biomedical articles,() but how clinicians adapt these articles into practice is not straightforward. In addition, the validity of biomedical research has recently been under scrutiny.() Bias in publication with emphasis on sensational discoveries over reproducibility, non-acceptance of negative studies, and the academic pressure to publish have all contributed to the unreliability of biomedical research. One consequence is the “medical pendulum” phenomenon, which pertains to treatments or diagnostic tools considered beneficial one decade and later proven to be of no value, or worse, harmful. An example in critical care is the pulmonary artery catheter, which was widely adopted in the 1980s and early 1990s, but later losing favor after retrospective observational studies suggested no benefit and possible harm,() followed by prospective randomized trials confirming such finding.(,) And while clinical trials are best in inferring causality, they are not adept at demonstrating small effect size which is typical with most critical care intervention administered to a heterogeneous group of patients. Moreover, clinical trials typically exclude important subgroups (older patients, those with comorbidities): findings may not be generalizable to the real-world.
Because of the limitations of clinical trials including cost, many guidelines are supported by low-quality evidence.() A survey of the American College of Obstetricians and Gynecologists practice bulletins showed only 29% of recommendations were level A, “based on good and consistent scientific evidence”() while an appraisal of the clinical practice guidelines from the American College of Cardiology and American Heart Association found only 314 of 2,711 recommendations (11%) were based on high quality evidence.()
[…]
Search
Search in:
Search
Search in:
Translating medical research into clinical practice guidelines is not trivial. There has been a surge in the number of published biomedical articles,() but how clinicians adapt these articles into practice is not straightforward. In addition, the validity of biomedical research has recently been under scrutiny.() Bias in publication with emphasis on sensational discoveries over reproducibility, non-acceptance of negative studies, and the academic pressure to publish have all contributed to the unreliability of biomedical research. One consequence is the "medical pendulum" phenomenon, which pertains to treatments or diagnostic tools considered beneficial one decade and later proven to be of no value, or worse, harmful. An example in critical care is the pulmonary artery catheter, which was widely adopted in the 1980s and early 1990s, but later losing favor after retrospective observational studies suggested no benefit and possible harm,() followed by prospective randomized trials confirming such finding.(,) And while clinical trials are best in inferring causality, they are not adept at demonstrating small effect size which is typical with most critical care intervention administered to a heterogeneous group of patients. Moreover, clinical trials typically exclude important subgroups (older patients, those with comorbidities): findings may not be generalizable to the real-world.
Because of the limitations of clinical trials including cost, many guidelines are supported by low-quality evidence.() A survey of the American College of Obstetricians and Gynecologists practice bulletins showed only 29% of recommendations were level A, "based on good and consistent scientific evidence"() while an appraisal of the clinical practice guidelines from the American College of Cardiology and American Heart Association found only 314 of 2,711 recommendations (11%) were based on high quality evidence.()
[...]
Comments