Events

Events

An Event is a figure that allows to homogenize the content of the output datasets. It consists, at least, of an object to be instantiated and a trait containing the necessary functions. In order to create an Event you first need to understand what it is. An event is an occurrence defined by having a patient identifier, category, start, group identifier, value, weight and end; of these only the first three are mandatory.The 7 elements needed to form an Event are:

• patientID: is the patient identifier.
• category: define the event's category.
• groupID: contains the ID of a group of related events.
• value: contains string values fot the molecule name, the diagnosis code, etc.
• weight: contains double values for medical acts weighting or other numerical values.
• start: is the start of event's period.
• end: is the end of event's period.

  patientID: String,
  category: EventCategory[A],
  groupID: String,
  value: String,
  weight: Double,
  start: Timestamp,
  end: Option[Timestamp] 

All events inherit AnyEvent and EventBuilder. AnyEvent is a trait with category value, and EventBuilder is a trait to built an Event.

ObservationPeriod and MedicalAct are good examples to illustrate the construction of an Event. The first one,ObservationPeriod, define the values patientID, category, start and if exist end. The other values are present but use default values. Its trait just assigns the value of the category and uses the apply method to build an Event.

 val category: EventCategory[ObservationPeriod] = "observation_period"

  /** Creates un Event object of type ObservationPeriod using a map function to map a dataset.
   *
   * @param patientID The value patientID from dataset.
   * @param start     The value start from dataset.
   * @param end       The value end from dataset.
   * @return Event[ObservationPeriod].
   */
  def apply(patientID: String, start: Timestamp, end: Timestamp): Event[ObservationPeriod] =
    Event(patientID, category, groupID = "NA", value = "NA", weight = 0D, start, Some(end))

And its object doesn't add up to anything, it just inherits the trait.

object ObservationPeriod extends ObservationPeriod

On the other side,MedicalAct defines values for all elements except end one. Its trait uses two apply methods to assign values according to need and not assign any value to category one, it's assigned in each object according to the type.

 
  override val category: EventCategory[MedicalAct]

  def apply(patientID: String, groupID: String, code: String, weight: Double, date: Timestamp): Event[MedicalAct] = {
    Event(patientID, category, groupID, code, weight, date, None)
  }

  def apply(patientID: String, groupID: String, code: String, date: Timestamp): Event[MedicalAct] = {
    Event(patientID, category, groupID, code, 0.0, date, None)
  }

The MedicalAct's object are various in accordance with the type of medical act. Their objects assign categories and in some cases have object that stores groupID values.

object BiologyDcirAct extends MedicalAct {
  override val category: EventCategory[MedicalAct] = "dcir_biology_act"

  object groupID {
    val PrivateAmbulatory = "private_ambulatory"
    val PublicAmbulatory = "public_ambulatory"
    val PrivateHospital = "private_hospital"
    val Liberal = "liberal"
    val DcirAct = "DCIR_act"
    val Unknown = "unknown_source"
  }
}