Recent work has advocated “cumulative damage modeling” (CDM) of passive component reliability in some contexts. Specifically, CDM may be useful in “Reliability Integrity Management” (RIM) (ASME Boiler and Pressure Vessel Code Section XI Division 2). CDM is an application of event-driven simulation performed to understand the interplay between component degradation mechanisms and various management strategies that may be implemented to manage the risk associated with passive component failure. One point of CDM is to model situations in which component functionality is lost only when a particular threshold of degradation is exceeded as a result of the cumulative effect of degradation mechanisms whose severity may fluctuate in time. Constant-failure-rate (e.g., Markovian) modeling has previously been applied to passive component reliability, but if threshold effects dominate in a particular context, there may be limitations to the applicability of a Markovian result in that context. The present paper compares traditional Markov modeling with CDM, with a view to (a) highlighting relative strengths and weaknesses of the two approaches, and (b) assessing what properties CDM needs to have – including what inputs are needed, and how to address uncertainty - to improve meaningfully on Markovian modelling. Probabilistic fracture mechanics (PFM) appears to be a special case of what we mean by CDM, and the existence of PFM work provides a useful way of benchmarking CDM approaches.