I believe a nursing diagnosis should drive the nursing process. You collect data (assessment), then define the patient's likely problem(s) (diagnosis), determine where the patient needs to be (goals) then design care to get them there (actions). Followed of course by evaluation and re-designing care based upon results.
How in the world can a NANDA statement drive clinical reasoning? I read student care plans daily. If I see the term "Ineffective breathing pattern" again I will absolutely poke sharp sticks in my eyes.
This "diagnosis" (ineffective breathing pattern) is equally suitable for the dying patient with Cheyne-Stokes respirations, the 18 year old trauma victim with a C3 cord transection, a 10 year old asthmatic who is ready to crash and burn, or an 80 year old post belly-surgery patient who won't use the incentive spirometer and is getting atelectasis. Puh-leeeze!
Why not just state what is wrong with a patient? How about
"Hypoxemia related to intrapulmonary shunt (secondary to pneumonia) AEB %sat 89 on Room Air, respiratory rate of 28, tachycardia and fever"
or something like that? It isn't a medical diagnosis and by golly, it immediately identifies the types of goals and actions that logically follow.
I have found NANDA to be an insurmountable obstacle to teaching students clinical reasoning. They work with their patient for the day, then that night, pull out their "nursing diagnosis" text, run their finger down a list of "diagnoses" and pick something that seems to fit the majority of their assessment findings. And from there, the care plan spirals down to a bunch of superficial blather.
Said enough. I'm sure I'm going to be flamed, but I just needed to vent.