I was reading through the material for one of my hospital’s mandatory CE courses when I came across this piece of information:
The accuracy of medication histories in hospital medical records came under scrutiny in a study that assessed 122 patients over the age of 65. The study compared the written medical record to a history obtained directly from the patient. The medication analysis focused on prescription medications and not over-the-counter medications. Overall, 60 percent of patient records had at least one error and 18 percent had three or more errors (an error was defined as either failure to record use of a medication or recording a medication that was not used).
Interesting. I do not doubt that medication errors aplenty have happened within our healthcare system: nurses and RTs are chronically understaffed and overworked. Floor nurses with six or seven patients can not realistically provide excellent care to all of their patients while doing all of the things they are mandated to do. RTs with twenty or thirty patients can not do an adequate job of effectively assessing and treating all of their patients in a timely manner. Errors happen. Documentation is flawed and easily forged.
However, I have a problem with asking patients to verify what has happened to them. In many cases I am sure patients remember clearly what happened to them: a huge number of patients are lucid and coherent. But even a lucid, coherent patient is going to have a hard time remembering exactly which of their thirty pills was given to them by whom at what time of day. A lucid patient with a lot going on emotionally, who was tired because of their hospital stay, may still give inaccurate but well-intentioned data. Factor into that the fact that a massive number of patients in the hospital are mentally ill, incompetent, illucid, demented, just plain angry, or actively psychotic, and you have the makings of some pretty inaccurate reporting. For example, if I walk into a room and a patient complains that they haven’t had a treatment in days, and then they say the EXACT SAME THING when I walk in again four hours later, how accurate are they? Not very.
Plus, even lucid patients may have hidden motives to report one way or another. Some people want to report what they know the surveyor wants to hear. Some people are angry and will obscure the truth or lie, claiming they were maltreated in order to make the hospital look bad.
The only time I would trust a patients statements about the care they received would be when they were communicating emotionally, e.g. “you guys treated me real good,” or when a patient is a Documentor, one of those guys who records everything that happens in a little spiral notebook. And while some Documentors are patients who are interested in their care and taking an active role in their recovery, some Documentors are angry old men who want to sue you. Differentiating can be difficult.
The bullet point to all this: any survey that relies on patient perception as opposed to hard data is going to be hugely flawed and, in my mind, immediately suspect. Asking patients about their opinions or perceptions is important so that we can improve their experience at the hospital; but when it comes to factual information, patient perceptions should be treated as inaccurate at best and as barefaced lies at worst. Medical errors happen, no doubt about it. But instead of asking people what happened to them, we should rely more on verifiable facts to tell us what happened, using patient-given information only if the patients are lucid and capable of good recall.
Ultimately, to reduce medication errors the only system that will work is a system of bar-code scanners that scan patients, then administrators, then meds. A pain in the tuckus to be sure, but a necessary evil.
I was reading through the material for one of my hospital’s mandatory CE courses when I came across this piece of information:
Interesting. I do not doubt that medication errors aplenty have happened within our healthcare system: nurses and RTs are chronically understaffed and overworked. Floor nurses with six or seven patients can not realistically provide excellent care to all of their patients while doing all of the things they are mandated to do. RTs with twenty or thirty patients can not do an adequate job of effectively assessing and treating all of their patients in a timely manner. Errors happen. Documentation is flawed and easily forged.
However, I have a problem with asking patients to verify what has happened to them. In many cases I am sure patients remember clearly what happened to them: a huge number of patients are lucid and coherent. But even a lucid, coherent patient is going to have a hard time remembering exactly which of their thirty pills was given to them by whom at what time of day. A lucid patient with a lot going on emotionally, who was tired because of their hospital stay, may still give inaccurate but well-intentioned data. Factor into that the fact that a massive number of patients in the hospital are mentally ill, incompetent, illucid, demented, just plain angry, or actively psychotic, and you have the makings of some pretty inaccurate reporting. For example, if I walk into a room and a patient complains that they haven’t had a treatment in days, and then they say the EXACT SAME THING when I walk in again four hours later, how accurate are they? Not very.
Plus, even lucid patients may have hidden motives to report one way or another. Some people want to report what they know the surveyor wants to hear. Some people are angry and will obscure the truth or lie, claiming they were maltreated in order to make the hospital look bad.
The only time I would trust a patients statements about the care they received would be when they were communicating emotionally, e.g. “you guys treated me real good,” or when a patient is a Documentor, one of those guys who records everything that happens in a little spiral notebook. And while some Documentors are patients who are interested in their care and taking an active role in their recovery, some Documentors are angry old men who want to sue you. Differentiating can be difficult.
The bullet point to all this: any survey that relies on patient perception as opposed to hard data is going to be hugely flawed and, in my mind, immediately suspect. Asking patients about their opinions or perceptions is important so that we can improve their experience at the hospital; but when it comes to factual information, patient perceptions should be treated as inaccurate at best and as barefaced lies at worst. Medical errors happen, no doubt about it. But instead of asking people what happened to them, we should rely more on verifiable facts to tell us what happened, using patient-given information only if the patients are lucid and capable of good recall.
Ultimately, to reduce medication errors the only system that will work is a system of bar-code scanners that scan patients, then administrators, then meds. A pain in the tuckus to be sure, but a necessary evil.