How to create multiple subsets of df based on main column/category?

Question

Let's suppose I have a very large df of these columns with sub-columns. Row 0 is the main column(it has many duplicates, Row 1 is the sub-column inside the column. I want to break each column into its' own df of sub-columns.

The only way that I know how to do this is by taking it apart one by one and joining them; however, there are close to 6,000 columns. Also, Patient ID needs to be the "Unique Identifier" in each df.

0	Benefits	Benefits	id	Care	Care
1	Medicare TYPE	Med Type	Patient ID	Type of care	position of care
2	Medicare A	Medicare B	Patient/51792geyfg	Skilled	primary
3	Not enrolled	UNKNOWN	Patient/rgyrebhrbv	MD	secondary

Desired Output: 1st df

Benefit's df

0	Medicare TYPE	Med Type
1	Medicare A	Medicare B
2	Medicare A	Medicare B
3	Not enrolled	UNKNOWN

2nd df ID df

0	Patient ID
1	Medicare A
2	Not enrolled

3rd df Care df

0	Type of care	position of care
1	Skilled	primary
2	MD	secondary