To create a relative time filter, you need to perform two steps:
Select Relative Time as the filter type: A relative time filter allows you to filter your data based on a relative time period, such as last week, next month, or current year. You can select Relative Time as the filter type from the drop-down menu of the filter bar.
Select Relative Time as the filter type and set the Relative To property: The Relative To property allows you to specify the reference point for your relative time period. For example, if you want to filter your data for the last quarter relative to today, you can set the Relative To property to Today. You can set the Relative To property from the filter dialog box after selecting Relative Time as the filter type. Reference: [Oracle Help Center], [Oracle Help Center]

Balance tree is the correct relational topology to model the patient dimension for a healthcare provider in Oracle Analytics Cloud. A balance tree is a type of schema that consists of one or more dimension tables that are related to each other by foreign keys. A balance tree schema allows you to model a dimension that has multiple attributes that are organized into different levels of detail. For example, you can model a patient dimension that has attributes such as contact information, primary physician, insurance information, and billing details. Each attribute can be stored in a separate table that is linked to a parent table by a foreign key. A balance tree schema enables you to perform drill-down analysis on your dimension without having to merge all the tables into one. The other types of schemas, such as distributed tree, universe, and star, are not suitable for modeling a patient dimension with multiple related tables. A distributed tree schema consists of one or more dimension tables that are not related to each other by foreign keys. A universe schema consists of one or more fact tables that are related to each other by foreign keys. A star schema consists of one fact table that is related to multiple dimension tables by foreign keys. Reference: [Oracle Help Center], [Oracle Help Center]

The step that is required in a data flow to create a new curated data set is Add Data. Add Data is the first step in any data flow that allows you to select one or more data sources to use in your data flow. You can choose from various types of data sources, such as databases, files, applications, or existing data sets. You can also specify how to join or union multiple data sources if needed. Reference: Oracle Analytics Cloud - Data Visualization User's Guide, [Oracle Analytics Cloud - Data Visualization User's Guide]

There are two possible ways to design the data flow and load data for these two separate data sets in data storage:
Create two separate data flows for the Country and Sales data sets. Then call one data flow from another using a Call Data Flow step. This will allow you to execute one data flow after another in a single run.
Create data flows for the Country and Sales data sets. Then create a sequence that defines the order of execution of these data flows using a Sequence step. This will allow you to run multiple data flows sequentially or in parallel. Reference: [Oracle Analytics Cloud - Data Visualization User's Guide], [Oracle Analytics Cloud - Data Visualization User's Guide]

Dashboard prompts are filters that allow users to dynamically change the data displayed in a dashboard. Some statements that are true about dashboard prompts are:
Prompt values can be limited based on values selected in other prompts. This is called cascading prompts, where the values of one prompt depend on the selection of another prompt.
Repository variables and session variables can be used in prompt default values. This allows users to set dynamic default values for prompts based on variables defined in the repository or the session. Reference: Oracle Analytics Cloud - Data Visualization User's Guide, Oracle Analytics Cloud - Data Visualization User's Guide