Katon Minhas
73e43f1c65
Refactor/update claude models * Revert prompts * Remove qa_qc tests * Remove valid lines * model aliasing * Haiku alias * remove erroneous import Approved-by: Alex Galarce
731 lines
25 KiB
Python
731 lines
25 KiB
Python
import ast
|
|
import hashlib
|
|
import os
|
|
import re
|
|
from datetime import datetime
|
|
from functools import cache
|
|
|
|
import pandas as pd
|
|
|
|
import src.prompts.investment_prompts as investment_prompts
|
|
import src.utils.llm_utils as llm_utils
|
|
from constants.delimiters import Delimiter
|
|
from src.prompts.investment_prompts import FieldSet, invoke_derived_term_date
|
|
from src.utils import string_utils
|
|
import src.config as config
|
|
|
|
# Determine the base directory for the project
|
|
BASE_DIR = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
|
|
# Path to the mappings folder
|
|
MAPPINGS_DIR = os.path.join(BASE_DIR, "crosswalk", "mappings")
|
|
|
|
|
|
def rename_columns(df):
|
|
"""
|
|
Use the provided mapping to rename columns in the data
|
|
|
|
Args:
|
|
df (pd.DataFrame): The input DataFrame with columns to be renamed.
|
|
Returns:
|
|
pd.DataFrame: The DataFrame with renamed columns.
|
|
"""
|
|
# Check if the DataFrame is empty, return it as is if it is
|
|
if df.empty:
|
|
return df
|
|
|
|
# If not empty, proceed with column renaming
|
|
rename_map = {
|
|
"PROCEDURE_CD": "CPT4_PROC_CD",
|
|
"PROCEDURE_CD_DESC": "CPT4_PROC_CD_DESC",
|
|
}
|
|
# Rename columns using the mapping
|
|
df = df.rename(columns=rename_map)
|
|
return df
|
|
|
|
|
|
def normalize_indicator_field(value: str) -> str:
|
|
"""
|
|
Normalize the indicator field value to 'Y' or 'N'.
|
|
|
|
Args:
|
|
value (str): The input value from an _IND field.
|
|
Returns:
|
|
str: 'Y' if the value is 'Y', otherwise 'N'.
|
|
"""
|
|
if isinstance(value, str) and value.strip().upper() == "Y":
|
|
return "Y"
|
|
return "N"
|
|
|
|
|
|
def format_rate_fields_with_commas(value: str) -> str:
|
|
"""
|
|
Format the rate with commas and two decimal places.
|
|
|
|
Args:
|
|
value (str): A string representing a numeric value.
|
|
Returns:
|
|
str: The formatted rate string.
|
|
"""
|
|
if string_utils.is_empty(value):
|
|
return ""
|
|
|
|
try:
|
|
numeric_value = float(value)
|
|
return f"{numeric_value:,.2f}"
|
|
except ValueError:
|
|
return ""
|
|
|
|
|
|
def remove_hyphens(value: str) -> str:
|
|
"""
|
|
Remove hyphens from the input string.
|
|
|
|
Args:
|
|
value (str): The input string.
|
|
Returns:
|
|
str: The cleaned string with hyphens removed.
|
|
"""
|
|
if not isinstance(value, str):
|
|
return ""
|
|
return value.replace("-", "")
|
|
|
|
|
|
def flatten_singleton_string_list(list_str: str) -> str:
|
|
"""Take in a list within a string, and if it has a single element, return that element.
|
|
If it has multiple elements, return the list as a string.
|
|
|
|
Args:
|
|
list_str (str): A string representation of a list.
|
|
|
|
Returns:
|
|
str: The single element if the list has one element, otherwise the list as a string.
|
|
"""
|
|
if string_utils.is_empty(list_str):
|
|
return ""
|
|
try:
|
|
# Convert the string to a list using ast.literal_eval
|
|
list_obj = ast.literal_eval(list_str)
|
|
# Check if list_obj is actually a list
|
|
if not isinstance(list_obj, list):
|
|
return list_str
|
|
# If the list has one element, return that element
|
|
if len(list_obj) == 1:
|
|
return list_obj[0]
|
|
# Otherwise, return the list as a string
|
|
return ", ".join(list_obj)
|
|
except (ValueError, SyntaxError):
|
|
return list_str
|
|
|
|
|
|
def validate_and_reformat_date(date_str: str) -> str:
|
|
"""
|
|
Validate if a string is in YYYY/MM/DD format or reformat it to YYYY/MM/DD if possible.
|
|
|
|
Args:
|
|
date_str (str): The input date string.
|
|
|
|
Returns:
|
|
str: The reformatted date string if valid or reformatted, otherwise the original string.
|
|
"""
|
|
if not isinstance(date_str, str):
|
|
return date_str
|
|
|
|
try:
|
|
# Check if the date is already in YYYY/MM/DD format
|
|
datetime.strptime(date_str, "%Y/%m/%d")
|
|
return date_str
|
|
except ValueError:
|
|
# Attempt to reformat the date from known formats
|
|
known_formats = ["%Y-%m-%d", "%m/%d/%Y", "%d-%b-%Y", "%d/%m/%Y"]
|
|
for fmt in known_formats:
|
|
try:
|
|
return datetime.strptime(date_str, fmt).strftime("%Y/%m/%d")
|
|
except ValueError:
|
|
continue
|
|
return date_str
|
|
|
|
|
|
@cache # memoize repeated calls to this function
|
|
def date_postprocessing(date: str) -> str:
|
|
"""Processes dates using `investment_prompts.date_fix_prompt`. Following the call to the LLM, the answer is extracted from the response and returned.
|
|
|
|
Args:
|
|
date (str): Input date string to be processed.
|
|
|
|
Returns:
|
|
str: Processed date; either a date in YYYY/MM/DD format, a duration (e.g. "3 years"), or "N/A".
|
|
"""
|
|
if string_utils.is_empty(date):
|
|
return "N/A"
|
|
|
|
prompt = investment_prompts.DATE_FIX_PROMPT(date)
|
|
response = llm_utils.invoke_claude(prompt, "haiku_latest", "date_fix")
|
|
return string_utils.extract_text_from_delimiters(response, Delimiter.PIPE)
|
|
|
|
|
|
def date_postprocess(df, field_json_path):
|
|
"""
|
|
Postprocess the date fields in the DataFrame.
|
|
"""
|
|
# Date field postprocessing using FieldSet definition
|
|
date_fields = FieldSet(file_path=field_json_path).filter(format="date")
|
|
for field in date_fields.fields:
|
|
if field.field_name in df.columns:
|
|
df[field.field_name] = list(map(date_postprocessing, df[field.field_name]))
|
|
|
|
# Derived termination date
|
|
df["AARETE_DERIVED_TERMINATION_DT"] = list(
|
|
map(
|
|
invoke_derived_term_date,
|
|
df["AARETE_DERIVED_EFFECTIVE_DT"],
|
|
df["TERMINATION_DT"],
|
|
)
|
|
)
|
|
|
|
return df
|
|
|
|
|
|
def normalize_auto_renewal_term(text: str) -> str:
|
|
"""
|
|
Normalize the auto-renewal term to a standard format.
|
|
|
|
Args:
|
|
text (str): The input auto-renewal term.
|
|
|
|
Returns:
|
|
str: The normalized auto-renewal term.
|
|
"""
|
|
if not text or not isinstance(text, str):
|
|
return ""
|
|
|
|
# Convert to lowercase and strip whitespace
|
|
text = text.lower().strip()
|
|
|
|
# Remove numbers enclosed in parentheses (e.g., "(1) ", "(12) ")
|
|
text = re.sub(r"\(\d+\) ", "", text)
|
|
|
|
# Define replacements
|
|
replacements = {
|
|
r"\bone\b": "1",
|
|
r"\btwo\b": "2",
|
|
r"\bthree\b": "3",
|
|
r"\btwelve\b": "12",
|
|
r"year to year": "1 year",
|
|
r"12 months": "1 year",
|
|
r"\btwelve months\b": "1 year",
|
|
r"\bone year\b": "1 year",
|
|
r"\bone-year\b": "1 year",
|
|
r"\b1 year\b": "1 year",
|
|
r"\bmonth to month\b": "1 month",
|
|
}
|
|
|
|
# Apply replacements
|
|
for pattern, replacement in replacements.items():
|
|
text = re.sub(pattern, replacement, text)
|
|
|
|
# Remove any remaining parentheses or extra spaces
|
|
text = text.strip().replace("(", "").replace(")", "")
|
|
|
|
text = text.replace("-", " ")
|
|
if text == "year":
|
|
text = "1 year"
|
|
|
|
return text
|
|
|
|
|
|
def normalize_cpt_fields(value):
|
|
"""
|
|
Normalize code values to a list format.
|
|
|
|
Args:
|
|
value (Any): Input value which could be a list, string, or other types.
|
|
|
|
Returns:
|
|
list: A normalized list of string codes.
|
|
"""
|
|
if not value or (isinstance(value, float) and pd.isna(value)): # Handle NaN values
|
|
return "" # Empty string
|
|
|
|
# First normalize to a list
|
|
if isinstance(value, (list, tuple)): # If already a list or tuple
|
|
result = [str(v).strip() for v in value]
|
|
elif isinstance(value, str): # If it's a string
|
|
value = value.strip()
|
|
if value.startswith("[") and value.endswith(
|
|
"]"
|
|
): # String representation of a list
|
|
try:
|
|
# Parse it first
|
|
parsed_list = ast.literal_eval(value)
|
|
result = []
|
|
# Process each item in the parsed list
|
|
for item in parsed_list:
|
|
item_str = str(item).strip()
|
|
# If the item contains commas, split it into multiple items
|
|
if "," in item_str:
|
|
result.extend([str(v).strip() for v in item_str.split(",")])
|
|
else:
|
|
result.append(item_str)
|
|
except (ValueError, SyntaxError):
|
|
result = [value] # If parsing fails, treat it as a single value
|
|
elif "," in value: # If it's a comma-separated string
|
|
# Split by comma and create a list
|
|
result = [str(v).strip() for v in value.split(",")]
|
|
elif "-" in value:
|
|
# Range format
|
|
result = [value]
|
|
else: # Single value
|
|
result = [value]
|
|
elif isinstance(value, (float, int)): # If it's a number
|
|
result = [str(int(value)).strip()]
|
|
else:
|
|
result = [str(value).strip()] # Default case for other types
|
|
|
|
return str(result) # Return the list directly
|
|
|
|
|
|
def generate_reimb_ids(df: pd.DataFrame) -> pd.DataFrame:
|
|
"""Generates REIMB_ID and REIMB_LESSER_OF_ID for each row in the DataFrame based on
|
|
the FILE_NAME and other fields.
|
|
The REIMB_ID is constructed using the following format:
|
|
{FILE_NAME}_exh_pg_{EXHIBIT_PAGE}_{index}_{hash_value}
|
|
|
|
the REIMB_LESSER_OF_ID is just the {hash_value} itself.
|
|
|
|
Where:
|
|
- FILE_NAME is the name of the file (without extension).
|
|
- EXHIBIT_PAGE is the page number of the exhibit.
|
|
- index is the index of the row in the group.
|
|
- hash_value is the first 8 characters of the MD5 hash of a concatenated string of relevant fields.
|
|
The relevant fields are SERVICE_TERM, REIMB_TERM, PROGRAM, PRODUCT, NETWORK, and LOB.
|
|
|
|
Args:
|
|
df (pd.DataFrame): Input DataFrame containing the columns FILE_NAME, EXHIBIT_PAGE, SERVICE_TERM, REIMB_TERM, PROGRAM, PRODUCT, NETWORK, and LOB.
|
|
The FILE_NAME column should contain the name of the file (with or without extension).
|
|
The EXHIBIT_PAGE column should contain the page number of the exhibit.
|
|
|
|
Returns:
|
|
pd.DataFrame: DataFrame with generated REIMB_IDs and REIMB_LESSER_OF_IDs for each row.
|
|
"""
|
|
# First sort to ensure consistent ordering
|
|
df = (
|
|
df.sort_values(by=["FILE_NAME", "EXHIBIT_PAGE"])
|
|
if "EXHIBIT_PAGE" in df.columns
|
|
else df.sort_values(by=["FILE_NAME"])
|
|
)
|
|
|
|
# Create a temporary copy to avoid SettingWithCopyWarning
|
|
df_temp = df.copy()
|
|
|
|
# Process each file+exhibit_page group to reset counter for each exhibit page
|
|
group_cols = (
|
|
["FILE_NAME", "EXHIBIT_PAGE"] if "EXHIBIT_PAGE" in df.columns else ["FILE_NAME"]
|
|
)
|
|
for group_key, indices in df.groupby(group_cols).groups.items():
|
|
if isinstance(group_key, tuple):
|
|
filename, exhibit_page = group_key
|
|
exhibit_page = str(exhibit_page).zfill(3)
|
|
else:
|
|
filename = group_key
|
|
exhibit_page = "000"
|
|
clean_filename = str(filename).split("/")[-1].replace(".", "_")
|
|
|
|
for i, idx in enumerate(indices):
|
|
row = df.loc[idx]
|
|
|
|
# Get fields to include in hash
|
|
service_term = str(row.get("SERVICE_TERM", ""))
|
|
reimb_term = str(row.get("REIMB_TERM", ""))
|
|
program = str(row.get("PROGRAM", ""))
|
|
product = str(row.get("PRODUCT", ""))
|
|
network = str(row.get("NETWORK", ""))
|
|
lob = str(row.get("LOB", ""))
|
|
|
|
# Create hash string
|
|
hash_string = f"{clean_filename}_{service_term}_{reimb_term}_{program}_{product}_{network}_{lob}"
|
|
hash_value = hashlib.md5(hash_string.encode()).hexdigest()[:8]
|
|
|
|
# Generate REIMB_ID and REIMB_LESSER_OF_ID
|
|
df_temp.at[idx, "REIMB_ID"] = (
|
|
f"{clean_filename}_exh_pg_{exhibit_page}_{i+1:03d}_{hash_value}"
|
|
)
|
|
df_temp.at[idx, "REIMB_LESSER_OF_ID"] = hash_value
|
|
|
|
return df_temp
|
|
|
|
|
|
def reorder_columns(df: pd.DataFrame, column_order: list[str]) -> pd.DataFrame:
|
|
"""
|
|
Reorders the columns of the DataFrame based on the given column order.
|
|
Steps:
|
|
1. Adds any missing columns from `column_order` to the DataFrame at once, filled with empty strings.
|
|
2. Reorders the columns of the DataFrame to match `column_order`.
|
|
3. Appends any columns in the DataFrame that are not in `column_order` to the end.
|
|
|
|
Args:
|
|
df (pd.DataFrame): The input DataFrame.
|
|
column_order (list[str]): The desired column order.
|
|
|
|
Returns:
|
|
pd.DataFrame: The reordered DataFrame.
|
|
"""
|
|
# Create a copy to avoid fragmentation
|
|
df_copy = df.copy()
|
|
|
|
# Find missing columns from column_order
|
|
missing_columns = [col for col in column_order if col not in df_copy.columns]
|
|
|
|
# Add all missing columns at once using a dictionary
|
|
if missing_columns:
|
|
# Create a dictionary of empty columns
|
|
empty_cols = {col: [""] * len(df_copy) for col in missing_columns}
|
|
# Add all missing columns at once with pd.concat
|
|
df_copy = pd.concat(
|
|
[df_copy, pd.DataFrame(empty_cols, index=df_copy.index)], axis=1
|
|
)
|
|
|
|
# Reorder columns to match column_order
|
|
ordered_columns = [col for col in column_order if col in df_copy.columns]
|
|
# Add columns in df that are not in column_order to the end
|
|
remaining_columns = [col for col in df_copy.columns if col not in column_order]
|
|
# Combine the ordered columns and remaining columns
|
|
final_column_order = ordered_columns + remaining_columns
|
|
|
|
# Return the DataFrame with reordered columns
|
|
return df_copy[final_column_order]
|
|
|
|
|
|
def auto_renewal(df: pd.DataFrame) -> pd.DataFrame:
|
|
# Normalize the 'AUTO_RENEWAL_TERM' column
|
|
if "AUTO_RENEWAL_TERM" in df.columns:
|
|
df["AUTO_RENEWAL_TERM"] = df["AUTO_RENEWAL_TERM"].apply(
|
|
normalize_auto_renewal_term
|
|
)
|
|
return df
|
|
|
|
|
|
def update_termination_date_for_conditions(df: pd.DataFrame) -> pd.DataFrame:
|
|
"""
|
|
Updates the 'AARETE_DERIVED_TERMINATION_DT' column based on specific conditions
|
|
involving 'TERMINATION_DT', 'AUTO_RENEWAL_IND', and 'AARETE_DERIVED_TERMINATION_DT'.
|
|
|
|
Args:
|
|
df (pd.DataFrame): The input DataFrame.
|
|
|
|
Returns:
|
|
pd.DataFrame: The updated DataFrame with modified 'AARETE_DERIVED_TERMINATION_DT' values.
|
|
"""
|
|
|
|
if (
|
|
"TERMINATION_DT" in df.columns
|
|
and "AUTO_RENEWAL_IND" in df.columns
|
|
and "AARETE_DERIVED_TERMINATION_DT" in df.columns
|
|
):
|
|
# Replace Y with 9999/12/31
|
|
df.loc[df["AUTO_RENEWAL_IND"] == "Y", "AARETE_DERIVED_TERMINATION_DT"] = (
|
|
"9999/12/31"
|
|
)
|
|
|
|
# Replace "N" and "year to year" with 9999/12/31
|
|
df.loc[
|
|
(df["TERMINATION_DT"] == "year to year")
|
|
& (df["AUTO_RENEWAL_IND"] == "N")
|
|
& (df["AARETE_DERIVED_TERMINATION_DT"] == "N/A"),
|
|
"AARETE_DERIVED_TERMINATION_DT",
|
|
] = "9999/12/31"
|
|
|
|
return df
|
|
|
|
|
|
def standardize_reimb_method_and_fee_schedule(
|
|
df: pd.DataFrame, VALID_UNIT_OF_MEASURE: list
|
|
) -> pd.DataFrame:
|
|
"""
|
|
Standardizes the 'AARETE_DERIVED_REIMB_METHOD' column and updates the
|
|
'AARETE_DERIVED_FEE_SCHEDULE' column based on specific values.
|
|
|
|
Args:
|
|
df (pd.DataFrame): The input DataFrame.
|
|
|
|
Returns:
|
|
pd.DataFrame: The updated DataFrame with standardized columns.
|
|
"""
|
|
if (
|
|
"AARETE_DERIVED_REIMB_METHOD" in df.columns
|
|
and "AARETE_DERIVED_FEE_SCHEDULE" in df.columns
|
|
):
|
|
# Standardize 'AARETE_DERIVED_REIMB_METHOD' values
|
|
# Copy AWP, ASP, WAC to AARETE_DERIVED_FEE_SCHEDULE column for the corresponding row
|
|
medrx_fee_schedule_values = ["AWP", "ASP", "WAC"]
|
|
df.loc[
|
|
df["AARETE_DERIVED_REIMB_METHOD"].isin(medrx_fee_schedule_values),
|
|
"AARETE_DERIVED_FEE_SCHEDULE",
|
|
] = df["AARETE_DERIVED_REIMB_METHOD"]
|
|
|
|
# Convert AARETE_DERIVED_REIMB_METHOD values [AWP, ASP, WAC] to 'MedRx'
|
|
df["AARETE_DERIVED_REIMB_METHOD"] = df["AARETE_DERIVED_REIMB_METHOD"].replace(
|
|
{"AWP": "MedRx", "ASP": "MedRx", "WAC": "MedRx"}
|
|
)
|
|
|
|
if "AARETE_DERIVED_REIMB_METHOD" in df.columns:
|
|
|
|
# Remove rows where 'AARETE_DERIVED_REIMB_METHOD' is 'DOFR' but 'REIMB_TERM' does not contain 'DOFR'
|
|
mask = (df["AARETE_DERIVED_REIMB_METHOD"] == "DOFR") & (
|
|
~df["REIMB_TERM"].str.contains("DOFR", na=False)
|
|
)
|
|
df = df[~mask]
|
|
|
|
# standardize the 'UNIT_OF_MEASURE' column
|
|
if "UNIT_OF_MEASURE" in df.columns:
|
|
df["UNIT_OF_MEASURE"] = df["UNIT_OF_MEASURE"].str.replace(
|
|
"Per Encounter", "Per Visit"
|
|
)
|
|
|
|
# remove cases of Per Member Per Month (PMPM)
|
|
df = df[~df["UNIT_OF_MEASURE"].isin(["Per Member Per Month (PMPM)"])]
|
|
|
|
# Set UNIT_OF_MEASURE as 'Per Unit' for Flat Rate reimbursement method with empty UNIT_OF_MEASURE
|
|
df.loc[
|
|
(df["AARETE_DERIVED_REIMB_METHOD"] == "Flat Rate")
|
|
& (df["UNIT_OF_MEASURE"].apply(string_utils.is_empty)),
|
|
"UNIT_OF_MEASURE",
|
|
] = "Per Unit"
|
|
|
|
# Ensure UNIT_OF_MEASURE is one of the valid values
|
|
df["UNIT_OF_MEASURE"] = df["UNIT_OF_MEASURE"].apply(
|
|
lambda x: x if x in VALID_UNIT_OF_MEASURE else ""
|
|
)
|
|
|
|
return df
|
|
|
|
|
|
def process_patient_age_range(df):
|
|
"""
|
|
Process the 'PATIENT_AGE_RANGE' column into 'PATIENT_AGE_MIN' and 'PATIENT_AGE_MAX'.
|
|
If a single number is provided, both min and max are set to that number.
|
|
Remove PATIENT_AGE_RANGE column.
|
|
|
|
Args:
|
|
df (pd.DataFrame): The input DataFrame.
|
|
|
|
Returns:
|
|
pd.DataFrame: The updated DataFrame with 'PATIENT_AGE_MIN' and 'PATIENT_AGE_MAX' columns.
|
|
"""
|
|
if "PATIENT_AGE_RANGE" in df.columns:
|
|
try:
|
|
# Create temporary columns for processing
|
|
df["PATIENT_AGE_MIN"] = df["PATIENT_AGE_RANGE"]
|
|
df["PATIENT_AGE_MAX"] = df["PATIENT_AGE_RANGE"]
|
|
|
|
# Handle hyphenated ranges
|
|
mask = df["PATIENT_AGE_RANGE"].str.contains("-", na=False)
|
|
if mask.any():
|
|
temp_split = df.loc[mask, "PATIENT_AGE_RANGE"].str.split(
|
|
"-", expand=True
|
|
)
|
|
df.loc[mask, "PATIENT_AGE_MIN"] = temp_split[0]
|
|
df.loc[mask, "PATIENT_AGE_MAX"] = temp_split[1]
|
|
|
|
except Exception as e:
|
|
# If any error occurs, keep the original values
|
|
pass
|
|
|
|
# Remove the original PATIENT_AGE_RANGE column
|
|
df.drop(columns=["PATIENT_AGE_RANGE"], inplace=True, errors="ignore")
|
|
|
|
return df
|
|
|
|
|
|
def add_aarete_derived_amendment_num(df: pd.DataFrame) -> pd.DataFrame:
|
|
"""
|
|
Add the 'AARETE_DERIVED_AMENDMENT_NUM' column to the DataFrame.
|
|
|
|
Args:
|
|
df (pd.DataFrame): The input DataFrame.
|
|
|
|
Returns:
|
|
pd.DataFrame: The updated DataFrame with the 'AARETE_DERIVED_AMENDMENT_NUM' column.
|
|
"""
|
|
|
|
if "CONTRACT_AMENDMENT_NUM" in df.columns:
|
|
df["AARETE_DERIVED_AMENDMENT_NUM"] = (
|
|
df["CONTRACT_AMENDMENT_NUM"]
|
|
.str.split("-")
|
|
.str[1]
|
|
.fillna(df["CONTRACT_AMENDMENT_NUM"])
|
|
)
|
|
|
|
return df
|
|
|
|
|
|
def add_aarete_derived_product(df):
|
|
"""
|
|
Add AARETE_DERIVED_PRODUCT column to df, based on PRODUCT column.
|
|
|
|
The new column should be Title Case of the PRODUCT column value.
|
|
"""
|
|
if "PRODUCT" in df.columns:
|
|
df["AARETE_DERIVED_PRODUCT"] = df["PRODUCT"].str.title()
|
|
return df
|
|
|
|
|
|
def update_lob_for_duals(answer_dicts):
|
|
final_answer_dicts = []
|
|
# Iterate through each row in the dataframe
|
|
for answer_dict in answer_dicts:
|
|
# Get SERVICE_TERM value
|
|
service_term = answer_dict.get("SERVICE_TERM", "").lower()
|
|
# Check if the SERVICE_TERM contains "Medicare" and "Medicaid" - case insensitive
|
|
if "medicare" in service_term and "medicaid" in service_term:
|
|
prompt = investment_prompts.DUAL_LOB_CHECK(service_term)
|
|
claude_answer_raw = llm_utils.invoke_claude(
|
|
prompt, model_id="legacy_sonnet", filename="", max_tokens=200
|
|
)
|
|
claude_answer_extracted = string_utils.extract_text_from_delimiters(
|
|
claude_answer_raw, Delimiter.PIPE
|
|
)
|
|
answer_dict["AARETE_DERIVED_LOB"] = claude_answer_extracted
|
|
final_answer_dicts.append(answer_dict)
|
|
return final_answer_dicts
|
|
|
|
|
|
def fill_empty_reimb_pct_rate(df):
|
|
"""
|
|
Fill 'REIMB_PCT_RATE' based on reimbursement method:
|
|
- Set 'REIMB_PCT_RATE' to '100' for rows where:
|
|
- 'REIMB_PCT_RATE', 'REIMB_FEE_RATE', and 'REIMB_CONVERSION_FACTOR' are all empty.
|
|
- Set 'REIMB_PCT_RATE' to 'N/A' for rows where:
|
|
- 'AARETE_DERIVED_REIMB_METHOD' contains specific grouper.
|
|
"""
|
|
required_cols = [
|
|
"REIMB_PCT_RATE",
|
|
"REIMB_FEE_RATE",
|
|
"REIMB_CONVERSION_FACTOR",
|
|
"AARETE_DERIVED_REIMB_METHOD",
|
|
]
|
|
|
|
if all(col in df.columns for col in required_cols):
|
|
# Fill with '100' when all three rate fields are empty
|
|
pct_empty = string_utils.is_empty(df["REIMB_PCT_RATE"])
|
|
fee_empty = string_utils.is_empty(df["REIMB_FEE_RATE"])
|
|
conversion_empty = string_utils.is_empty(df["REIMB_CONVERSION_FACTOR"])
|
|
|
|
fill_mask = pct_empty & fee_empty & conversion_empty
|
|
df.loc[fill_mask, "REIMB_PCT_RATE"] = "100"
|
|
|
|
# Set to N/A for specific reimbursement methods
|
|
na_methods = ["Grouper", "Per Diem", "Flat Rate", "Per Visit", "Case Rate"]
|
|
|
|
na_mask = df["AARETE_DERIVED_REIMB_METHOD"].isin(na_methods)
|
|
df.loc[na_mask, "REIMB_PCT_RATE"] = "N/A"
|
|
|
|
return df
|
|
|
|
|
|
def remove_redundant_reimb_info(df):
|
|
"""
|
|
Checks if "REIMB_" fields are the same as their corresponding 1:1 fields.
|
|
If they are the same, replace the "REIMB_" values with ""
|
|
|
|
Args:
|
|
df (pd.DataFrame): The input DataFrame.
|
|
|
|
Returns:
|
|
pd.DataFrame: The updated DataFrame with redundant reimbursement field values removed.
|
|
"""
|
|
COLUMN_PAIRS = [
|
|
("REIMB_EFFECTIVE_DT", "AARETE_DERIVED_EFFECTIVE_DT"),
|
|
("REIMB_TERMINATION_DT", "AARETE_DERIVED_TERMINATION_DT"),
|
|
("REIMB_PROV_NAME", "PROV_GROUP_NAME_FULL"),
|
|
]
|
|
for column_tuple in COLUMN_PAIRS:
|
|
reimb_col = column_tuple[0]
|
|
one_to_one_col = column_tuple[1]
|
|
if all(col in df.columns for col in [reimb_col, one_to_one_col]):
|
|
mask = df[reimb_col] == df[one_to_one_col]
|
|
df.loc[mask, reimb_col] = ""
|
|
return df
|
|
|
|
|
|
def deduplicate_provider_columns(df: pd.DataFrame) -> pd.DataFrame:
|
|
"""Deduplicate provider fields in the DataFrame.
|
|
|
|
This function removes PROV_GROUP_* values from the PROV_OTHER_* columns if they are already present.
|
|
In addition, it deduplicates PROV_OTHER_* values.
|
|
|
|
E.g. If PROV_GROUP_TIN is "123456789" and PROV_OTHER_TIN is "123456789 | 987654321 | 987654321",
|
|
then PROV_OTHER_TIN will be updated to "987654321".
|
|
|
|
|
|
Args:
|
|
df (pd.DataFrame): The input DataFrame.
|
|
|
|
Returns:
|
|
pd.DataFrame: The updated DataFrame with deduplicated provider fields.
|
|
"""
|
|
provider_columns = [
|
|
"PROV_GROUP_TIN",
|
|
"PROV_GROUP_NPI",
|
|
"PROV_GROUP_NAME_FULL",
|
|
"PROV_OTHER_TIN",
|
|
"PROV_OTHER_NPI",
|
|
"PROV_OTHER_NAME_FULL",
|
|
]
|
|
other_columns = ["PROV_OTHER_TIN", "PROV_OTHER_NPI", "PROV_OTHER_NAME_FULL"]
|
|
|
|
# Check if all provider columns exist in the DataFrame
|
|
if all(col in df.columns for col in provider_columns):
|
|
# Process each row
|
|
for index, row in df.iterrows():
|
|
# Get GROUP values for this row
|
|
group_tin = str(row.get("PROV_GROUP_TIN", "")).strip()
|
|
group_npi = str(row.get("PROV_GROUP_NPI", "")).strip()
|
|
group_name = str(row.get("PROV_GROUP_NAME_FULL", "")).strip()
|
|
|
|
# Simple mapping of OTHER columns to their corresponding GROUP values
|
|
column_mappings = {
|
|
"PROV_OTHER_TIN": group_tin,
|
|
"PROV_OTHER_NPI": group_npi,
|
|
"PROV_OTHER_NAME_FULL": group_name,
|
|
}
|
|
|
|
# Process each OTHER column
|
|
for other_col, group_val in column_mappings.items():
|
|
other_value = str(row.get(other_col, "")).strip()
|
|
|
|
if (
|
|
other_value
|
|
and not string_utils.is_empty(other_value)
|
|
and other_value != "UNKNOWN"
|
|
):
|
|
# Split by pipe and strip whitespace
|
|
other_list = [
|
|
v.strip() for v in other_value.split("|") if v.strip()
|
|
]
|
|
|
|
# Remove corresponding GROUP value if present
|
|
if (
|
|
group_val
|
|
and not string_utils.is_empty(group_val)
|
|
and group_val != "UNKNOWN"
|
|
):
|
|
other_list = [v for v in other_list if v != group_val]
|
|
|
|
# Remove duplicates while preserving order
|
|
seen = set()
|
|
deduped_list = []
|
|
for item in other_list:
|
|
if (
|
|
item not in seen
|
|
and not string_utils.is_empty(item)
|
|
and item != "UNKNOWN"
|
|
):
|
|
seen.add(item)
|
|
deduped_list.append(item)
|
|
|
|
# Update the DataFrame
|
|
df.at[index, other_col] = (
|
|
" | ".join(deduped_list) if deduped_list else ""
|
|
)
|
|
return df
|