The sequential use of alternative treatments for chronic conditions represents a complex, dynamic intervention pathway; previous treatment and patient characteristics affect both choice and effectiveness of subsequent treatments. Evidence synthesis methods that produce the least biased estimates of treatment-sequencing effects are required to inform reliable clinical and policy decision-making. A comprehensive review was conducted to establish what existing methods are available, outline the assumptions they make, and identify their shortcomings.

The review encompassed both meta-analytic techniques and decision-analytic modelling, any disease condition, and any type of treatment sequence, but not diagnostic tests, screening, or treatment monitoring. It focused on the estimation of clinical effectiveness and did not consider the impact of treatment sequencing on the estimation of costs or utility values.

The review included ninety-one studies. Treatment-sequencing is usually dealt with at the decision-modelling stage and is rarely addressed using evidence synthesis methodology for clinical effectiveness. Most meta-analyses are of discrete treatments, sometimes stratified by line of therapy. Prospective sequencing trials are scarce. In their absence, there is no single best way to evaluate treatment sequences, rather there is a range of approaches, each of which has advantages and disadvantages and is influenced by the evidence available and the decision problem. Due to the scarcity of data on sequential treatments, modelling studies generally apply simplifying assumptions to data on discrete treatments. A taxonomy for all possible assumptions was developed, providing a unique resource to aid the critique of decision-analytic models.

The evolution of network meta-analysis in HTA demonstrates that clinical and policy decision-making should account for the multiple treatments available for many chronic conditions. However, treatment-sequencing has yet to be accounted for within clinical evaluations. Economic modelling is often based on the simplifying assumption of treatment independence. This can lead to misrepresentation of the true level of uncertainty, potential bias in estimating the effectiveness and cost effectiveness of treatments and, eventually, the wrong decision.