In this section, we're going to talk about balancing workload with quality when it comes to case definitions and public health surveillance. We can think about case definitions as being sensitive or specific. Let me tell you more about what that means. So, a sensitive case definition is one in which you're unlikely to miss any true cases but on the other hand, you may be more likely to capture patients who are not true cases. So, sensitive case definitions are very inclusive but you may find a lot of patients who don't really have the disease that you're interested in. Let me show you what this looks like. Here on the right side, you see a population of patients, they're in your health facility, let's say, and the patients with blue faces are those which are true cases. The patients with peach colored faces are those which are not true cases. So, typically, if your case definition is sensitive, you're going to be capturing most of those true cases. It'll look something like this. So, all the faces on the red box are the ones who would capture if you had a sensitive case definition. So, you can see here that you haven't missed any of the true cases, but you have captured a lot of the not true cases with your case definition. Importantly, when we use case definitions we often don't know what the underlying truth here is, all we know is that these patients have met our case definition. So, this is more of a concept to help me think about since it's sensitive case definitions versus specific. So, let's talk about what a specific case definition might look like and how it's different from the sensitive case definition. So, a specific case definition may miss some true cases, but you're less likely to capture patients who are not true cases. So, what does this mean? What does this look like? Well, it would look something like this. So, you can see here that the specific case definition missed one of the true cases that you weren't included in that case definition in the red box, but almost all the patients that were included were true cases. It's a bit more efficient if you want to use a specific case definition. But for some diseases you want to be really, really sensitive because missing a case could be an important public health missed opportunity. So, when you're thinking about sensitive versus specific case definitions, you need to think about your particular disease and the objective of your surveillance system. If it's really important for your surveillance system not to miss any true cases, then you need to go sensitive. If it's really important that you capture only true cases, then you need to go specific. So, sensitivity and specificity in case definitions are always a bit of a trade-off, and you have to decide which way you want to go based on their surveillance objective. We previously mentioned suspect probable and confirmed cases. Some surveillance systems use only syndromic case definitions instead of laboratory confirmation for their case definitions. So, syndromic surveillance case definitions are those which are based only on signs and symptoms of the patient. As you can imagine, these are more sensitive than specific. Influenza like illness is a great example of this. This is an internationally recognized syndromic surveillance case definition. It's fever and cough, with or without sore throat. This is an easy to use definition, is very sensitive and it's cheap, which means you can use it everywhere. Again, draw from surveillance systems all around the world. But the limitations are it captures many people who do not truly have influenza. So, again, there's a trade-off here between what's easy and what's very specific. Other surveillance systems use laboratory-based case definitions. This all depends on your surveillance objective. For a laboratory-based case definition, testing is required for confirmation but usually it's linked to a syndrome. So, if you recall before we talked about suspect cases who may have samples collected and then those are tested to see if you confirm whether the case is a true case or not. Well, in influenza they often do the same thing. So, some influenza surveillance programs require laboratory confirmation. In those cases, often patients who have influenza-like illness will have a nasopharyngeal swab collected and tested by real-time RT-PCR to see if they can find RNA from influenza viruses in the respiratory tract to the patient. If they do, they would call this a lab confirmed case. The strengths of this case definition are that it increases our confidence in our measurement. We feel much more confident saying that this patient has influenza than a patient who just has influenza-like illness, for example. But there are limitations, it's always a trade-off. This type of surveillance case definition is more expensive. It creates time lags in your system because you have to wait for the diagnostic test results to come back to know whether or not you have a case and sometimes this can take days or weeks. There are also limitations of the diagnostic test. Every laboratory test has limitations. So, you have to understand really what the limitations of your test are to know truly how confident you are in the test results that you're using. So, whether you choose syndromic or laboratory confirmed for your case definition, will depend on your surveillance objective and really what you're after as well as the resources you have. Something else to consider, is whether or not your surveillance system will have active or passive case finding. Whether the system is active or passive is a characteristic of that system, it's defined based on who is identifying the cases. If the public health authority is actively looking for cases and finding them, then we would call that system active. If the public health authority is receiving reports from practitioners, we would call that passive. Now, it's important to remember here, active or passive is not based on action. There's action in both of these systems. Rather it's who is actively looking for the cases. So, in an active system, the public health authority is the one that's expanding the resources to go out and find the cases. While in a passive system, someone else is actively looking for the cases and the public health authorities are just requesting or sometimes even mandating that they be reported. In reality, this is a bit of a continuum and some public health surveillance systems can have elements of both passive and active. Let me give you a brief example. So, let's say that physicians in a city are asked and required by public health authorities to report all cases of tuberculosis. But the public health authorities even though they're not actively looking for the cases may send people out to meet with physicians to encourage reporting, they may be following up very actively with physicians and with health care providers to make sure that they're reporting all the cases they find. So, they may be very active. So, that kind of system would probably be classified as passive but more on the active side. So, it's not always one or the other, but you should know what these concepts mean. So, surveillance systems aren't always one or the other, active or passive. They can have elements of both. But it's important to know what the concepts of active and passive mean. Remember, it's not based on action. They all have action. It's based on who is actively identifying the cases.
