One of the numerous plates that the government is juggling at the moment is plans to increase health surveillance in the United States. I’m viewing this with mixed feelings, because I think that there are some really good things about it, and also some potentially really bad things, and it’s worth exploring them in a bit more detail.
Basically, the idea is that we need to compile better health statistics. We need a better reporting system in place for health care providers to submit and record information, we need better tracking systems to follow people and monitor health outcomes, we need better mechanisms for identifying and addressing public health risks. We also need to be able to use this information in meaningful and helpful ways. All of this together could theoretically improve access to health services, quality of health services, and health outcomes.
Nations with ‘socialised’ medicine have pretty excellent surveillance systems in place, thanks to the highly centralised nature of their health care systems. When people are using unified electronic medical records, receiving care through government providers, and participating in a standardised system, it makes them easier to track. There’s also an incentive for strong health surveillance because it saves money in the long term. Evidence-based medicine, in addition to being better for patients, is also usually cheaper to implement. Consequently, maintaining accurate and detailed statistics is critical.
One of the biggest advantages to increased surveillance that I can see is that we could start generating a lot more statistics and using them to really drill down into the disparities in the United States. This is dependent on collecting the right kind of data, though. For example, studies seem to suggest that transgender people experience significant barriers when it comes to accessing health care, and that our health outcomes tend to be less positive than those of cisgender people. This is something that could be tracked through health surveillance, if anyone bothered to enquire about gender identification and if detailed statistics were collected.
This requires generating systems that can be used to accurately ask about gender and record responses. Having a system that identified cis men, cis women, trans men, and trans women would be a good start, but it would leave out nonbinary people and nongender people. We would be lumped into one of these categories, even though we wouldn’t belong in it, and this would mean that statistics about us would be hard to find and track. It would be impossible to tell, for example, how many nonbinary people with ovaries develop ovarian cancer, or how many nongender people with prostates get prostate cancer. Not being able to track this information accurately could mean that risk factors and opportunities for intervention are missed.
Thinking that we are collecting accurate and detailed data can be dangerous, just as not collecting gender-based statistics at all can be dangerous. Because people may believe that they are collecting all relevant material, that the health surveillance system is not letting people slip through the cracks, that we are collecting valuable and important material, and something major could be missed.
There are tremendous class and race disparities in this country that could also be tracked through a good health surveillance system, but, again, the system is only as good as the data collection. Using absolute targets for class, for example, could mean that people who are technically middle class, but living in areas with a high cost of living, aren’t reported on accurately. Likewise, if there is no box someone can check for ou racial identification, it means that no statistics will be available, and, again, risk factors could be missed. There’s only so much that you can correct with formulas; ultimately, you enter a realm where populations are not being reported on because data is not being collected for them, or it is, but it is not being classified in the right way. This means that disparities will be missed, just as they are already being missed right now.
This does not mean that we should not try. Obviously, collecting some data is better than collecting no data. Admitting that there may be shortcomings in that data and trying to find ways to improve it is critically important too, though. If we can’t talk about the holes that might be present in the data we are collecting, we are going to have trouble identifying at risk populations to target with additional studies.
Let’s not forget, too, that sometimes increased health surveillance results in more health interventions, and this can actually have a negative impact on outcomes. Here in the United States, there’s a prevailing attitude that more intervention is always better, when it comes to health care, that we are ‘the best in the world’ when it comes to health care because we will go to great lengths in every case, but studies actually show that this isn’t always borne out.
When interventions are provided for conditions that might clear up on their own, that’s costly. To the patient, the doctor, the system. Interventions can put people at risk. Risks need to be weighed and considered. Having better health surveillance will help people evaluate those risks, as it will be possible to compare different treatments and different outcomes, but it could also result in health policy that might be harmful. Like deciding that because [Condition Y] is very common in [Population Z], we should screen aggressively for it in that population, or possibly even treat prophylactically, even if the evidence doesn’t actually support that. What if Condition Y rarely kills people, and most people who have it die of other causes? What if it presents no real hardships? What if there’s really no reason to treat everyone who has Condition Y, only people with a specific form of it? What if it makes sense to take a watchful waiting approach to see if treatment is needed or not?
Health, as we know, is not one size fits all. It’s important that as we increase health surveillance and collect more statistics, that we remember that individuals are not statistics, and statistics are not the only predictors of health outcomes, not when there are so many factors involved.