March 10, 2014

The variance among prefectures in the long term insurance system in Japan

So What ?

Japan has the system of long-term insurance for elderly people.

People who want to use the system have to apply the certification for it.

After the certification, they are classed as several levels.

Prefectures operate the system.

Although the process of the certification is standardized, stakeholders of it are various (e.g. doctors, investigators and other specialists in healthcare).

Then I'm afraid that changing the number of the certification would varied among prefectures.

Today, I will try to check it.

Here's the result.

So What ?

As I mentioned before, there seems to be the variance among prefectures.

The difference between max and min are about ten percent each year.

Moreover, you can see a plunge from 2006 to 2010 in the level “要介護1” (In fact, there is no public data from 2007-2009.)

It is because the system was changed in 2006 and the a part of “要介護1” was redivided into “要支援”.


youkaigo <- read.csv("", = TRUE)
youkaigo$date <- as.Date(youkaigo$date)
youkaigo$prop <- round(youkaigo$prop, 1)/100
youkaigo$area <- factor(youkaigo$area, levels = 1:6, labels = c("北海道/東北", 
    "関東", "甲信越/北陸", "東海/近畿", "中部/四国", "九州/沖縄"))
youkaigo <- reshape(youkaigo[, -5], idvar = c("保険者", "area", "date"), 
    timevar = "variable", direction = "wide")
colnames(youkaigo) <- gsub("prop\\.", "", colnames(youkaigo))

# visualize
n1 <- nPlot(要支援 ~ date, group = "保険者", data = youkaigo, type = "lineChart")
n1$xAxis(tickValues = "#![11323,11700,12060,12420,12790,13150,13520,13880,14250,14615,14980,15345,15710,16071,16430]!#", 
    tickFormat = "#!function(d) {return d3.time.format('%Y')(new Date( d * 86400000 ));}!#")
n1$yAxis(tickFormat = "#!function(d) {return d3.format('%')(d)}!#")
n1$chart(forceY = "#![0]!#",
         useInteractiveGuideline=TRUE )
n1$addControls("y", value = "要支援", values = names(youkaigo)[-(1:3)])
n1$show("iframesrc", cdn = TRUE)

No comments:

Post a Comment