How filters work ?

Wiki - 2016-02-26

Table of contents

1. What is the purpose of filters ?

2. How to create a filter ?

3. Combine filters together

4. Unactivate or edit filters

5. Save time with Historic

6. Go even Faster with Quick filters

7. Precisions on operators

    Is between

    Is unknown (NC)

    Is known (not NC)

    Is not

Filters is one of the most important function of the site

They allow you to select subgroups of patients among your series

What is the purpose of filters ?

Instead of studying all of your patients at once, you can decide to study only a subset of your patients, for example women, people older than 90 years, right-handed patients, ...
Once a filter is activated, it will remain active everywhere on EasyMedStat so you can list patients, get descriptive statistics or even perform statistical tests on you subset of patients.

How to create a filter ?

  1. Go to Filters (Fig. 1)
  2. Choose a variable to filter and click on the "Ok" button (Fig. 2)
    For example, if you want to filter patients older than 90 years, choose "Birth Date" in the list
  3. Choose an operator which can be "Equals", "Is between", ... (Fig 3)
  4. Choose values you want to filter (Fig. 3)
    Specify minimum and maximum birth dates to select patients according to their age
  5. Click on "Add this filter" to activate your filter

Fig. 1 - Click the "Filters" link to create a new filter

Fig. 2 - Choose a variable to filter

Fig. 3 - Choose values for you filter
You can specify a minimum value, a maximum value or both

Combine filters together

Let's say you want to analyze women over 70 y.o. whose follow-up is more than 2 years. Very easy !
  1. Add a filter on "Gender" > "Equals" > "Woman"
  2. Add a filter on "Birth Date" > "Is Between" > 1947-01-01 and +∞
  3. Add a filter on "Follow-up" > "Is between" > 24 and +∞

Once your 3 filters have been created, you can display them by clicking on "Active filters" (Fig. 4)

When several filters are added, they are combined with the AND operator

So, the more you add filters, the less you select patients

Fig. 4 - By clicking on "active filters", you can see which filters are currently active

Unactivate or edit filters

  1. Go to "Active filters" to see the filters currently applied to your series.
  2. The Trash button allows you to remove a filter
  3. The Edit button allows you to modify a filter without removing it (changing the values or the operator)

Save time with Historic

There are two ways of accessing filter historic :
  1. When you click on the link "Filters" or when you create a new filter, last filters created are displayed below the filter creation form (Fig. 1)
  2. On "Patients" page, and in statistics pages, you can find a Select List (Fig. 5 and 6) which displays the 50 last filters created. You just need to choose a filter and the page will reload, with the filter activated !

Fig. 5 - Filter Select List

Fig. 6 - Filter Select List : choose one filter among the 50 last filters created

Go even Faster with Quick filters

At many places on the website, you can find this small button which allows you to quickly create a new filter.
This button will display filters that are related to the value next to the button (Fig. 7). Just give it a try to understand ...

Fig.7 - Quick Filter on the value 63.55

Precisions on operators

Is between
When you use filters on numeric or event variables, you can filter with two different operators :
  • Is between will look for values ≤ or ≥ to the boundaries you choose
  • Is strictly between will look for values strictly inside the boundaries, and excluding them (operators < >)

You can let one of the boundaries empty to find values that are superior or inferior to a certain number.
It is not possible yet to mix one operator ≤ or ≥ with an operator < or > (to be coming ...)

Is unknown (NC)
You want to find patients whose value is unknown for a variable ? Use this operator (Fig. 8). You don't have to choose a value for this operator.

Is known (not NC)
It is the opposite of "Is unknown" (Fig. 8). It returns all patients who have a value for the variable, whatever this value is.
NB : number of patients Not NC + number of patients NC = total number of patients

Is not
Selects all the patients that have a value for the variable which is different from the one you select. E.g. Patients whose blood type is not O+. (Fig. 8)
It means that "NC" patients will not be selected : since we don't know their value, we can not affirm that this value is not something ...

Fig. 8 - Example operators for a List value

Recent posts

Most peer-review journals will ask you a "Statistical analysis" chapter in your article (in Material and Methods), in which you can quote EasyMedStat as follows: "Statistical analysis was performed with the online software EasyMedStat (www.easymed ... Read more
It's easy to customize a bar chart. Read more
You can easily perform a survival analysis, with the Kaplan-Meier method. Read more
Variables can be grouped into categories and subcategories to help you. Read more
To add a patient, click on the "Patients" menu and then choose "+ Add patient" Read more

Copyright EasyMedStat© 2013

Easy Made Stat - Association Loi 1901 - n° W922009408
Neuilly-sur-Seine - France

Contact us - Legal terms