Mongodb time series query - mongodb

Question

I'm new to MongoDB coming from a relational world. Any help would be greatly appreciated. I have a list of patients with the following attributes:
{"id": 101,
"demographics": {"sex": "male","dob": "1/1/1984"},
"hospital_visits": [
{ "date": "1/1/2012",
"diagnosis": ["diabetes","fractured hip"]
},
{ "date": "3/1/2012",
"treatment": ["hip replacement"],
"outcome": "normal discharge"
},
{ "date": "5/1/2012",
"diagnosis": ["hip infection"],
"outcome": "inpatient admission"
}
]
},
{"id": 102,
"hospital_visits": [
{ "date": "1/1/2013",
"diagnosis": ["fractured hip"]
},
{ "date": "3/1/2013",
"treatment": ["hip replacement"],
"outcome": "normal discharge"
}
]
}
Now, if I want to find out the number of patients that had a "fractured hip" as a diagnosis at some point (in one of the hospital visits), had a "hip replacement" treatment done subsequently at a later date (not in the same hospital visit), and within 3 months from the day of "hip replacement" treatment had an "inpatient admission" as an outcome, how do I frame my MongoDB query? Going by this logic, in the given example data, clearly patient 101 is a hit while patient 102 is not, so the count is 1. But just how do I formulate this query in MongoDB? Any ideas? Would it be easier if I change the document structure to address this question?
Many thanks!

Answer 1

One option can be:
If these queries are in considerable in number, then, whenever you update the document check for the condition that you want, and then store a suitable variable ( say some Boolean ) and you can index that variable.
The count will return very fast in that case.