Since Clinians are often working with big data that has been obscured for HIPAA laws to protect PII and PHI, how do you suppose an evaluator could successfully evaluate the two qualitative questions you posed? This question is for the whole group.
So these are the questions used as an example by Young:
• Have patients’ quality of life improved following treatment?
• Is patient participation in our treatment program “responsible” for their improvement?
Desarae asks how the evaluator evaluates the two above qualitative questions since patient information is highly protected and regulated.
One possible solution is to look at the aggregated data without any identifiers. This way the evaluator only looks at the bulk of the data not knowing which individual responded to each questions. Such approach has some drawbacks such as the inability to follow-up any responses with particular patients. — APC
My current work is in health and civilian solutions, specifically related to big data services and solutions. Most of our clients are government programs. It is usually very expensive for researchers to access this kind of data with extreme restrictions. That being said, I can think of other instances where programs are being created at a national or state level to evaluate the quality of health care by providers. This quality check is also being used to base payments for care givers.
A lot of these programs still include coded identifiers and back-end programs to identify fraud. Data for even one state or county often is counted in zettabytes, that is a lot of data. The most expensive iPhone, for example, can have up to 64 GBs of data.
Plus, all systems require codes of codes and encryption with varying levels of access requirements. I’m over simplifying and being fairly vague, but I’m trying to paint a picture of the restrictions health care researchers, statisticians, UX designers, program managers, clinicians, and program directors are working with when they work in a field related to human-centered data (especially with something as sensitive as health care data).
So how do we evaluate “quality of life improvement” in a meaningful and defensible way that can promote positive change? Is this criteria based on the wellness of the patient? Can we say if the patient is no longer sick or doesn’t have to return for more treatment, that maybe their quality of life has improved? That’s not really a fail safe though, what’s to say they just never returned to the doctors office or if they do maybe they forget to bring up that particular issue? How do we define quality of life (e.g. happiness/psychological factors, a doctor’s brilliance, a doctor’s bed side manner, a smiley face chart, a survey that could depend on their mood at the time, actual health improvements like the ability to walk again or full removal of a cancerous tumor that never resurfaces)?
How could the patient not be responsible for their own treatment? If a patient is not responsible for following up with treatment, who is (e.g. the government, a spouse/care-giver, their neighbors, a doctor, their insurance, their boss)? If the person is hospitalized or in a home care center, this may be easier to identify.