Abstract: Knowledge of ‘circadian phase’, or the timing of an individual’s internal 24 hour clock, is becoming increasing important in all aspects of medicine. Proper diagnosis of circadian rhythm sleep disorders requires a measurement of the circadian phase, to first of all define whether a disorder is present at all and then to provide a differential diagnosis of the circadian aspects of the disorder in relation to sleep timing – is there a disorder in the timing of the clock and sleep, or just sleep, for example, which is important in determining the possible cause and for treatment options? Once a diagnosis is made, the appropriate timing of chronobiotic therapies, such as light or melatonin, which can reset the clock either earlier or later depending on the circadian time of the treatment, also requires a circadian phase assessment. While clinical assessments of circadian phase have traditionally been viewed as a specialist requirement in sleep medicine, it has long been know that the circadian system influences virtually all areas of our physiology, behaviour and metabolism and therefore many non-sleep related clinical disorders. Knowledge of circadian phase would enhance the interpretation of many standard clinical tests, possibly changing diagnostic criteria based on the time the test was performed. It is also clear that the efficacy of drugs also varies by circadian phase but this information is rarely incorporated into practice. The widespread adoption of ‘chronomedicine’, an approach that would revolutionize personalized medicine, is limited by the methods available to measure circadian phase. Sleep timing, which has often been used due to the seeming ease of measurement, is a poor marker of the circadian system. Hormonal markers, which can be very accurate, often require controlled environments or specialized techniques, limiting their use. A number of new techniques are in development that utilize more sophisticated models for prediction of circadian phase or use ‘omics’ technology to try to determine phase using a unique combination of multiple parameters in one or two samples.