Audiqueen Analytics is a powerful module to query patients and extract test results based on several criteria and export them to Excel.  


The Patient Query is its key building block.


Set selection criteria for the query

The first search step consists of querying patients and results based on several criteria.

Audiqueen Analytics starts by selecting the data source (1), defining the patient selection criteria (2), the type of requested result (3), and the result details (4).


  • Data source: this is either the full database or a shortlist (in the form of a CSV file) of several patient IDs. Such a shortlist can be the result of a previous query on the EMR (electronic medical record) database.
    Read this again!  This means that you can build complex searches by making two or more consecutive queries.  The first one creates a shortlist and the second one gets the results you are looking for.  For instance: you can first make a shortlist of all patients who have unaided thresholds of X before cochlear implant surgery.  You then export only the patient IDs of this shortlist to a CSV file.  You then start a new query in this shortlist (by selecting it as the data source) and you now look for the speech audiogram 1 year after CI  surgery.
  • Patient Filter: subjects can be filtered on gender, date of birth, undergone (surgical) events, keywords, or type of hearing devices.
  • Results Filter: results can be filtered on result type, whether the result was obtained with or without hearing devices, and the result date. For the result date, different options exist:
    • Absolute date: in this case, the first or last result is chosen, or results within a certain time window; if this window contains multiple results, one can opt for only one result per subject, e.g. the earliest, the latest, the middle one.
    • Relative to birth date: in this case, an age window is chosen. if this window contains multiple results, one can opt for only one result per subject, e.g. the earliest, the latest, the middle one.
    • Relative to Event date (only possible if an event was used as a selection criterion): in this case, the first or last result before or after the event can be chosen, or all results within a certain time window, or, in case of multiple results in such window, only one of those.

The field Synopsis lets you query specific synopsis values.  

  • In the case of a numeric parameter, the value should be within the two values you enter.
  • In the case of a text parameter, the value should (exactly) match the string and/or strings that you enter

Wildcards 

Whenever fields are used to match text fields, matching means to be exactly the same.

If the original text field is "unreliable responses", entering "responses" in the query field will not cause a match!

Therefore, you may want to use wildcards.  The  following wildcard symbols are available:

  • % stands for any text 
  • _ stands for any single character
  • [... ] stands for any single character within the specified range, e.g. [a-f] or [abcmnq]
  • [^...] stands for any single character NOT within  the specified range, e.g. [^a-f] or [^abcmnq]

So the combination of these two search criteria ...

will look for notes containing the word 'respons' but not the character 'f'.  Notes containing the words 'false responses' will therefore not be selected.



Examples of queries on synopsis fields.  The left figure queries ABR peak V latencies between 6 and 10 msec.  
The right figure queries on Genetic GJB2 parameters containing the string "c.".

   
 

  • Curve Filter: this filter should hardly ever be used.  It is only meant to select special cases.  The available fields depend on the test type selected in the Results Filter.  Please note that this filter also affects the results that are exported to excel (see below).
    • Side: this restricts the results to only those marked as 'left', 'right', or 'both'. So in case you are only interested in results obtained from the left ear, here is the place to select this. If you check 'both', the query only retrieves results marked as binaural, hence with a binaural index displayed in the grid. If you are interested in results from all ears, right and left, do not use this filter. Audiqueen retrieves them by default.
    • Conduction: this restricts the results to only air, bone, or 'air or bone' conduction.  In most cases, it is not necessary to use this, because the index chosen in the Result Filter is already defined in terms of the conduction (see FAQ).
    • Stimulus: here you can restrict your query to results obtained with a specific stimulus, such as 'pulse' for audiometry.  If you add more than one stimulus, Audiqueen retrieves all results obtained with stimulus 1 OR with stimulus 2.
    • Frequency: this defines which thresholds will be exported to the excel sheet (see below).  Please make sure to leave all frequencies checked that are required to calculate the index chosen in the Results Filter.
    • Scoring: here you can restrict the output to excel to only item or segment scores.


The query criteria are summarized in the summary at the bottom of the window.

For instance, after having entered these criteria:


The summary says:


Preview subjects and results

Click on ‘Search’ to execute the query. This yields a shortlist of patients meeting the criteria.

In this example, Audiqueen has retrieved 29 results with an average score of 70%.  Click here to learn how Audiqueen calculates the average results.

The query can be refined, either by changing the criteria and running the query again or by unchecking certain subjects or results in the preview pane.

For each patient on the shortlist, all available test results can be inspected.

The red rectangle indicates all available test results in Audiqueen concerning this particular patient.


Here, the user selected speech audiometries which were obtained with a cochlear implant on the ipsilateral side. 
The list shows the dates, the speech list used, and the average score of each test.


Exclude fake patient records

It may occur that dummy patient records have been made in the past, for training purposes, exercise, etc.  You want to be sure that the results of these fake records do not corrupt your analytics.  For that matter, make sure to mark those patient records as fake.  See here how to do this.


Export the results to Excel

The results can be exported to excel for further processing.