Whilst the amount of research, funded and published, has grown enormously, there is little to suggest increased outputs have led to real improvements in patient care. Equally worrying, the growth and volume of evidence
1 has been accompanied by a corrosion in the quality of evidence,
2 which has compromised medicine’s ability to provide affordable, effective, high value care.
Disquiet about the lack of high quality evidence cannot be dismissed as the grumblings of a disgruntled few.
2 Serious systematic bias,
3 error,
3 and wastage
4 are now too well documented in all areas of medicine
5 and across the entire research ecosystem, from research to implementation.
Problems include: poorly managed commercial and academic vested interests,
6 bias in the research agenda, failure to take account of the patient perspective in research questions and outcomes,
7 poor research design,
4 the rising use of surrogate outcomes,
8 lack of transparency, lack of independent scrutiny, ghost authorship,
9 publication bias,
10 reporting bias
11 including under-reporting of harms,
12 trials that fail to follow their protocols
13 or are stopped too early,
14 and over-interpretation and misuse of the results of research,
15 uncorrected errors,
16 and undetected fraud.
17 5 These are just some ways in which our flawed system generates evidence that has reached the wrong result, or conclusions that don’t matter to real world patients.
18 studies suggest that most published research (including a range of study designs) is more likely to be false than true.
16
As medical research activity has intensified over the past 50 years, so too have these problems, aggravated by major structural problems with guideline production,
19 significant regulatory failings,
5 and significant delays in the withdrawal of harmful drugs.
20 All of this contributes to escalating costs of treatment,
21 medical excess,
22 and avoidable harm.
20 Add to this the unmanageable volume of evidence,
1 the prohibitive cost of doing research,
23 and the lack of evidence to support shared decision making,
24 and it is clear that we have a major structural problem with the current production and use of evidence.
2 If left unaddressed, these inherent problems may become entrenched and unsolvable.
