A couple of thoughts.
It's impossible to have 're-experiencing' symptoms in relation to an event that you never actually 'experienced' (that actually occurred) in real life in the first place. The more the anxiety/distress is about something that 'could have happened' (but didn't), it doesn't make any sense to me to consider this a 'traumatic event/ trauma' in one's past fueling current distress.
There's obviously a continuum and room for nuance (and debate) here. If a veteran served in Iraq in 2008 and was doing daily patrols in an area where there were constant enemy attacks (IED's, snipers, firefights, etc.) but that veteran was just 'lucky' enough never to have actually been shot at, blown up, or injured, I still would say that the veteran was clearly 'exposed to a warzone' sufficient for it to meet Criterion A for development of possible PTSD. However, other situations (to me) clearly don't meet criterion A. Some examples:
1) being stateside (never deployed) during the Vietnam era but being afraid that one would be deployed is not a 'Criterion A'
2) being in the Navy and serving on an aircraft carrier during normal operations in the Mediterranean sea and, at certain points, having to don/doff nuclear/biological/radiological protective gear (as part of routine, universally-experienced, precautionary training/ preparation drills) is not a 'Criterion A' event
There are others, but you get my point. There will ALWAYS be a requirement of clinical judgment (and careful interviewing) around the putative or potentially traumatic event (PTE) and if/how/why the individual experienced it as 'traumatic.' There is a subjective component to this and there always will be. They eliminated Criterion A2 from the DSM-5 (after DSM-IV) because they were largely considering events (originally) that anyone would find traumatic (i.e., canonical traumas like getting shot, getting raped, or seeing someone turn to mist in front of your eyes). However, in cases that are at the borders or fringes of Criterion A, to me, it is absolutely essential for the clinician to interview the patient around the question of what, specifically, did you experience as 'traumatic' about that experience? A technical phrase that is a handy way of stating your doubt is that their presentation/ report of their history and symptoms does not align with known patterns of symptom expression (or etiopathology) of the condition in question or the known 'natural history or etiopathogenesis' of the disorder. Someone claiming nightly nightmares about terrorist attacks that disrupt 90% of their sleep 10 years after being 'exposed' to being deployed to Saudi Arabia for an objectively non-eventful deployment on a large base just...doesn't really make clinical sense as a trauma- and stressor-related disorder. The claimed symptoms are just ridiculously out of proportion to the putative stressor (to me). The more that the putative relationship between the claimed 'stressor' and current symptoms 'strains credulity,' the more likely you're dealing with factors other than the 'traumatic event' causing/fueling those present symptoms and the more you're likely dealing with other factors (e.g., malingering/feigning/response bias, personality disorder traits or conditions, substance use/withdrawal effects, etc.).
Another concern is quite practical. Our trauma-focused therapies (PE/CPT/EMDR) generally require focus on a 'most traumatic event' as part of how they are implemented. If someone is deployed to a large base in Saudi Arabia for three years (but no discrete traumatic event happened to them) but they are claiming that 'the entire deployment' was their 'trauma' because they were in constant fear of something happening (that didn't) how do you handle that as 'the event?' It's too diffuse and if someone is having crippling anxiety, nightmares, hyperarousal, etc. about an event that 'didn't happen' vs. did happen 10 years ago then you're probably less likely dealing with PTSD and more likely dealing with other diagnostic entities or clinical case formulations.
In cases where veterans are presenting with requests to be 'evaluated for PTSD' or they 'think they have PTSD' or 'everyone tells them they have PTSD' and they are reporting 'trauma histories' without clear traumatic events in them...I'd be cautious. Are they pushing 'trauma' as their problem (and PTSD as their diagnosis) or are you (as the clinician interviewing them and getting to know them) seeing a trauma- and stressor-related disorder pattern of symptomatology emerge organically from an exploration of their past events and current expressed patterns of thoughts/emotions/behaviors and symptoms? This is important because, to me, a veteran who clearly has PTSD doesn't have to put a whole lot of effort in to convice me that they have PTSD. If they are simply cooperating with the assessment procedures (chart review, interview, observations, self-report symptom checklists, maybe objective psychological testing), the data from that process are convincing me that PTSD is a real possibility or is the likely best way to conceptualize the case.
When you interview them in detail around their putative PTSD symptoms (endorsed on the PCL-5) by using CAPS-level interview questions...do their responses make sense? Do they even understand the questions? Recently I've been bemused(?) at how frequently I will ask a veteran 'how much' a particular symptom (e.g., intrusive traumatic memories or nightmares or emotional reactivity to cues) "bothers them" or "how much of a problem" the symptom is for them (straight from the CAPS-5) and they either respond with (a) bafflement / confusion / irritation at the question ('what kind of question is that? I don't understand the question') or they (b) respond with how frequent they are claiming to experience the symptom ("all the time...the memories never stop...all day"). You learn a lot from listening to their answers to open-ended questions. Asking someone who actually has PTSD 'how much' their nightmares bother them is not a trick question nor an inappropriate / weird question to ask. Those who respond in that way...are likely not experiencing PTSD and may be experiencing Service-Connection Deficit Disorder instead.
Is there a clear 'texture' and detail to their descriptions of their experience of these symptoms? You'll learn a lot by careful and detailed interviewing around their endorsed re-experiencing and avoidance symptoms. We also tend to forget that we are doing psychological evaluations. This means that a conversation about the meaning of the events (or deployment experiences) is crucial to understanding the nature of the psychopathology that may have developed (hypothetically) in relation to these events or experiences. There is a narrative element here. I see a lot of vets (who don't have PTSD but 'want' to be diagnosed with PTSD) coming in to clinic and simply 'citing traumas' and 'reciting symptoms (or symptom labels).' 'I was traumatized by my deployment to Iraq...I have night terrors, I'm anxious all the time...I have anxiety...I never sleep...I don't leave the house...I have PTSD from it' is the beginning of the interview process...not the end. It's not a process of I show up at a psychologist appointment and say (a) I have a history of potentially-traumatic events and (b) I can recite several labels for PTSD symptoms. It's a conversation. Get to know your client. Get a comprehensive understanding of their psychological history and other contributing factors to their current psychopathological symptoms (e.g., mood disorders, substance use disorders, extreme personality traits, overall adjustment/ maladjustment historically and now, etc.). The clinical picture will either make sense or not make sense and that takes time, data, and clinical hypothesis testing. Have them fill out some cognitive-behavioral self-monitoring forms where they write down disturbing events that occur in between appointments and their specific cognitive/emotional/behavioral responses to these events. What patterns emerge? There are lots of presentations these days of veterans who come in with all sorts of labels that imply that they're really suffering severe psychopathology but this may conflict with my observations of them or the history in their chart (or their occupational and interpersonal functioning). It helps to do psychological testing with the MMPI-2-RF or the PAI to get another source of data rather than the symptom self-report checklists like the PHQ/GAD-7/PCL which are almost universally elevated and practically useless in many cases.