Up to three quarters of hospital tests are not being followed up, suggests a systematic review of international evidence, published in BMJ Quality and Safety.
This failure can have serious implications for patients, including missed or delayed diagnoses and even death, the study shows.
The authors base their findings on a systematic review of evidence published in English between 1990 and 2010, and available on reputable research databases.
Analysis of the findings of the 12 studies which were suitable for inclusion indicated that between 20% and 61% of inpatient test results, and between 1% and 75% of tests on patients treated in emergency care, were not followed up after discharge.
Critical test results and results for patients moving between healthcare settings, such as from inpatient to outpatient care or to general practice, were most likely not to be pursued, the study showed.
Only two of the 12 studies described fully electronic test management systems, and the rate of missed results was high in both. But the authors point out that this might be because technology makes the issue more explicit and easier to measure.
And rates were just as high in paper-based systems and those using a mix of paper and electronic records.
In 2008, the World Alliance for Patient Safety identified poor test follow-up as one of the key processes leading to unsafe patient care, and the analysis of the seven studies looking at the impact on patients reveals a range of consequences.
These include missed or delayed diagnoses of infectious disease and cancer, inappropriate or unnecessary antibiotic prescriptions, and even death.
One study, which assessed clinical negligence claims, found that 79 of the 112 claims - almost two thirds - involved missed diagnoses in emergency care settings that ended up harming the patient.
In 13 of these 79 claims, the broken link in the chain occurred when the test results were either transmitted to, or received by, the care provider.
"There is evidence to suggest that the proportion of missed test results is a substantial problem, which impacts on patient safety," conclude the authors.
Explore further: Testing times: Detecting HIV in resource-limited settings