sentinel surveillance

Sentinel surveillance trades coverage for depth. Rather than asking every hospital or general practice in a country to report every case, a health agency recruits a smaller network of sites that agree to report in detail, often taking laboratory samples as well as counting patients. The European Centre for Disease Prevention and Control describes its influenza networks this way: the participating doctors typically represent between 1 and 5 percent of those working in a country or region, and they supply both clinical reports and specimens for testing.

The advantage is quality and speed. A small network can be asked for the kind of detail a national reporting system could never sustain, which is why sentinel systems are the usual way of tracking seasonal respiratory viruses. The cost is that the resulting numbers are not a national total. A sentinel count says what those sites saw, and trends over time are the thing it measures well.

Comparisons therefore depend on the network staying stable. When the same sites report across two seasons, counting the same way, the change between them is meaningful even though neither figure is a complete count. When sites join or drop out, the raw totals move for reasons that have nothing to do with the disease, which is why agencies publishing sentinel data usually state how many sites took part and for how long.