Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Follow up to my previous PR. merge that one first and then this when ready #6

Merged
merged 2 commits into from
May 22, 2023
Merged

Follow up to my previous PR. merge that one first and then this when ready #6

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b15ecbe
[ADDED] the following:
lpanda2 May 22, 2023
2240ee4
[ADDED] chronic condition charts
lpanda2 May 22, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
64 changes: 45 additions & 19 deletions pages/02_financial_summary.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -80,11 +80,23 @@ def pmpm_by_claim_type():
group by 1, 2
having sum(paid_amount) > 0
order by 1, 2 desc
), pharmacy_summary as (
select
year(dispensing_date)::text || '-' ||
lpad(month(dispensing_date)::text, 2, '0')
as year_month
, 'pharmacy' as claim_type
, sum(paid_amount) as paid_amount_sum
from core.pharmacy_claim
group by 1
), together as (
select * from spend_summary union all
select * from pharmacy_summary
)
select
*
, paid_amount_sum / member_month_count as paid_amount_pmpm
from spend_summary
from together
join pmpm._int_member_month_count using(year_month)
"""
data = util.safe_to_pandas(conn, query)
Expand Down Expand Up @@ -143,8 +155,8 @@ def pmpm_by_service_category_1_2():
return data


year_month_values = year_months()
year_values = sorted(list(set([x[:4] for x in year_month_values['year_month']])))
year_month_values = sorted(list(set(year_months()["year_month"])))
year_values = sorted(list(set([x[:4] for x in year_month_values])))

## --------------------------------- ##
## Header
Expand Down Expand Up @@ -190,18 +202,18 @@ def pmpm_by_service_category_1_2():
[["paid_amount_sum", "member_month_count"]].sum()\
.assign(paid_amount_pmpm = lambda x: x["paid_amount_sum"] / x["member_month_count"])

st.table(pmpm_claim_type_data)

# plost.bar_chart(
# data=pmpm_claim_type_data
# bar='',
# value=,
# stack='normalize',
# direction='horizontal',
# legend='top',
# height=200
# )

st.markdown("## Claim Type")
st.markdown("""
Explore the per member per month costs across different claim types to gain insights into healthcare expenditure patterns. Inpatient spend will tend to be much higher than professional spend. Dig deeper to find out what is hidden in these costs.
""")
plost.bar_chart(
data=pmpm_claim_type_data,
bar='claim_type',
value='paid_amount_pmpm',
direction='horizontal',
legend='top',
height=200
)


## --------------------------------- ##
Expand Down Expand Up @@ -235,7 +247,7 @@ def pmpm_by_service_category_1_2():
## Service Category 2
## --------------------------------- ##
service_cat_options = service_1_data["service_category_1"].drop_duplicates().tolist()
col1, col2 = st.columns(2)
col1, col2, col3 = st.columns(3)
with col1:
st.markdown("""
Use the following dropdown to get more detail on the service category that interested you.
Expand All @@ -246,22 +258,33 @@ def pmpm_by_service_category_1_2():
options=service_cat_options,
label_visibility='collapsed',
)
with col3:
selected_year_month = st.selectbox(
label="Select a Year Month",
options=["All Time"] + year_month_values,
label_visibility='collapsed',
)

service_2_data = pmpm_by_service_category_1_2()
service_2_data = service_2_data.loc[
service_2_data["year_month"].str[:4].isin(selected_range)
& (
(service_2_data["year_month"] == selected_year_month) |
(selected_year_month == "All Time")
)
& service_2_data["service_category_1"].isin([selected_service_cat])
].drop("service_category_1", axis=1).reset_index(drop=True)
service_2_data = service_2_data\
.groupby("service_category_2", as_index=False)\
[["paid_amount_sum", "member_month_count"]].sum()\
.assign(paid_amount_pmpm = lambda x: x["paid_amount_sum"] / x["member_month_count"])


service_2_chart = alt.Chart(service_2_data).mark_bar().encode(
x="paid_amount_pmpm",
y=alt.Y("service_category_2", sort="-x"),
tooltip=["service_category_2", "paid_amount_pmpm"]
y=alt.Y("service_category_2", sort="-x", axis=alt.Axis(labelLimit=300)),
tooltip=["service_category_2", "paid_amount_pmpm"],
).properties(
height=300
)

st.altair_chart(service_2_chart, use_container_width=True)
Expand All @@ -271,5 +294,8 @@ def pmpm_by_service_category_1_2():
## Cost Variables
## --------------------------------- ##
st.markdown("## Cost Variable Quality Summary")
st.markdown("""
Explore common descriptive statistics to gain a comprehensive understanding of the quality and distribution of a particular claim cost variable.
""")
cost_summary_data = cost_summary()
st.dataframe(cost_summary_data, use_container_width=True)
97 changes: 97 additions & 0 deletions pages/03_chronic_conditions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,97 @@
import streamlit as st
import pandas as pd
import altair as alt
import plost
import util
import components

conn = util.connection(database="dev_lipsa")

@st.cache_data
def year_months():
query = """
select distinct
year(claim_end_date)::text || '-' ||
lpad(month(claim_end_date)::text, 2, '0')
as year_month
, sum(paid_amount)
from core.medical_claim
group by 1
having sum(paid_amount) > 10
order by 1
"""
data = util.safe_to_pandas(conn, query)
return data


@st.cache_data
def pmpm_by_chronic_condition():
query = """
with conditions as (
select distinct
year(condition_date)::text || '-' || lpad(month(condition_date)::text, 2, '0') as year_month
, claim_id
, patient_id
, code
, condition
, condition_family
from core.condition
inner join chronic_conditions._value_set_tuva_chronic_conditions_hierarchy vs on condition.code = vs.icd_10_cm_code
where code_type = 'icd-10-cm'
)
, medical_spend as (
select
year(claim_start_date)::text || '-' || lpad(month(claim_start_date)::text, 2, '0') as year_month
, claim_id
, patient_id
, sum(paid_amount) as medical_paid_amount
from core.medical_claim
group by 1, 2, 3
), merged as (
select
year_month
, condition_family
, sum(medical_paid_amount) as medical_paid_amount_sum
from conditions
join medical_spend using(patient_id, claim_id, year_month)
group by 1, 2
)
select
*
from merged
join pmpm._int_member_month_count using(year_month)
order by 2, 1
"""
data = util.safe_to_pandas(conn, query)
return data


## --------------------------------- ##
## Chronic Condition
## --------------------------------- ##
st.markdown("## Chronic Condition")
st.markdown("""
Certain conditions will be more expensive in your population. Here we see that the top driver of spend
is `Cardiovascular disease`. The second highest driver is `Metabolic Disease`.
""")

year_month_values = year_months()
year_values = sorted(list(set([x[:4] for x in year_month_values['year_month']])))
selected_range = components.year_slider(year_values)


chronic_condition_data = pmpm_by_chronic_condition()
chronic_condition_data = chronic_condition_data.loc[
chronic_condition_data["year_month"].str[:4].isin(selected_range)
]
chronic_condition_data = chronic_condition_data\
.groupby("condition_family", as_index=False)\
[["medical_paid_amount_sum", "member_month_count"]].sum()\
.assign(medical_paid_amount_pmpm = lambda x: x["medical_paid_amount_sum"] / x["member_month_count"])\
.round()\
.sort_values("medical_paid_amount_pmpm", ascending=False)

st.dataframe(
chronic_condition_data[["condition_family", "medical_paid_amount_pmpm"]],
use_container_width=True
)
Loading