My answer to this question is paper charts (i.e. lack of EMR). Keeping patients' health information in huge, overflowing bulky charts *by provider* is insanely sloppy and inefficient... not to mention irresponsible, and even dangerous. When the patient moves, that information certainly doesn't move with the patient. When the patient gets a new doctor in a different state, s/he has to remember his/her own meds, record of immunizations, whatever. The new doc & the old doc COULD communicate, but typically do not. And typically the patient doesn't have easy access to his/her own info, either. So the typical scenario is that s/he starts all over with a brand new doc.
And this is the reason why America has such a shoddy record on immunizations compared to other industrialized nations. The record keeping is so spotty. This doesn't just affect discrete issues like immunizations etc. It also has a huge impact on care for patients with comorbid conditions (which is like, practically everyone). For example, the patient with chronic ailments like diabetes & depression ON TOP OF something else like coronary artery disease probably has multiple doctors... yet unless these docs are all part of an integrated system like Kaiser Permanente, more likely, they have nothing to do with each other and have no idea what other care the patient is receiving from the others, short of what the patient can remember himself/herself and think to regurgitate at appropriate times. This is not only dreadfully inefficient, it's downright irresponsible & dangerous. A functional EMR would go a long way towards improving these care-linkage deficiencies.
The other huge component an EMR would help is quality control/process improvement. Improvement isn't possible without analysis through studies such as a medical chart review. And ongoing medical chart reviews aren't likely devoid consistent, easily accessible data. I've done paper chart reviews myself and process is so tedious, the data is so inconsistent, it's almost impossible to get anywhere. With the adoption of an EMR, doing studies would be much quicker, not to mention more accurate. The ability to do regular, consistent studies is a fundamental pre-requisite to improving quality of care. Data should be stored consistently, and it should be stored by patient, following the patient around, rather than in bits & pieces by physician on bulky paper charts.
I work in a hospital with an EMR right now and cannot imagine working in one with paper charts... my whole work day would just be filled with so much extra work. The other great thing about the EMR we have is that staff DO use it to communicate to each other. Getting universal adoption of EMR is something that CAN be done in the near future which will dramatically improve the quality of care. So in contrast to the many other healthcare problems out there today (47 million uninsured, etc.), here's a HUGE problem that does have a pretty clear cut solution.