[Healthcare Research into Anti-VEGF Therapy: Selection and Methodological Precautions].

Health care research has emerged as an approach to assess and improve quality of care and patient outcomes in the real world. It also has the potential to reduce healthcare costs by providing evidence to guide healthcare decisions.Randomised controlled trials (RCTs) theoretically offer the ideal study design to support treatment decisions. In RCTs, randomisation (formal chance) determines treatment allocation, which prevents selection bias from distorting the parameters of anti-VEGF treatment effects. Despite this advantage, only a minority of patients qualify for inclusion in neovascular age-related macular degeneration or diabetic macular oedema trials, which limits the validity of the results to the whole patient population seen in clinical practice. The evidence base for anti-VEGF is deficient in terms of matching the characteristics of patients encountered in clinical practice, and a more representative sample of older people and those with significant disability must be included in future trials. RCTs often do not address other knowledge gaps, including treatment delay, comparisons for less frequent types of CNV, monitoring of rare or late toxicity events or systemic safety.Observational studies, or studies in which treatment allocation occurs independently of investigators' choice or randomisation, can complement RCTs by providing data that is more relevant to the circumstances under which intravitreal therapy is routinely practiced. However, it is important to be aware of the strengths and limitations of such observational study designs, in order to optimise the design as well as the analytic techniques. Selection bias and loss-to-follow-up cause make comprehensive interpretation and careful analysis necessary. Future reports should focus on time-to-event analysis, as this is much less prone to loss-to-follow-up and improves adherence of (functionally) one-eyed or good responders. Observational studies and pragmatic trials can test new hypotheses and possible license extensions. The bearing of RCT findings on day-to-day practice can then be assessed. The data can be interpreted in a more meaningful manner by practicing clinicians if evidence is integrated from a variety of different study designs and methodologies.