Modules

Main module.

comorbidipy.calculator.comorbidity(df: DataFrame, id: str = 'id', code: str = 'code', age: str = 'age', score: str = 'charlson', icd: str = 'icd10', variant: str = 'quan', weighting: str = 'quan', assign0: bool = True)[source]

Calculate Charlson and Elixhauser Comorbidity Scores from ICD codes

Parameters:

  • df (pd.DataFrame) – Pandas Dataframe with at least 2 columns for id and code

  • id (str, optional) – Name of column with unique identifier. This may be for a single patient or an episode. Defaults to “id”.

  • code (str, optional) – Name of column with ICD codes. Defaults to “code”.

  • age (str, optional) – Name of column with age. Defaults to “age”. If age is not provided, set this to None.

  • score (str, optional) – One of “charlson”, “elixhauser”. Defaults to “charlson”.

  • icd (str, optional) – One of “icd9”, “icd10” and descibes the version used in the code column. Defaults to “icd10”.

  • variant (str, optional) – Mapping variant to use. Defaults to “quan”.

  • weighting (str, optional) – Weighting variant to use. Defaults to “quan”.

  • assign0 (bool, optional) – Should the less severe form of a comorbidity be set to 0 if the more severe form is present. Defaults to True.

Raises:

  • KeyError – Raised if id or code are not in df.columns.

  • KeyError – If age is not None and age is not in df.columns.

  • KeyError – Raised if combination of score, icd and variant not found in mappings. Call comorbidipy.get_mappings() to see permitted combinations.

Returns:

Returns dataframe with one row per id. The dataframe will contain comorbidities in columns as well as a comorbidity_score column. If score`==”charlson” and `age is given, age_adjusted_comorbidity_score and survival_10yr are calculated as below.

age_adjusted_comorbidity_score = comorbidity_score + 1 point for every decade over 40 upto a maximum of 4 points

\[10yr survival = 0.983^(e^(0.9 * comorbidity_score))\]

Return type:

Pandas DataFrame

comorbidipy.calculator.disability(df: DataFrame, id: str = 'id', code: str = 'code') → DataFrame[source]

Identify disabilities and sensory impairments from ICD10 codes

Parameters:

  • df (pd.DataFrame) – Pandas dataframe containing at least id and code columns

  • id (str, optional) – Name of column containing patient identifier. Defaults to “id”.

  • code (str, optional) – Name of column containing ICD10 codes. Defaults to “code”.

Raises:

KeyError – Error is raised if id or code columns are not present in dataframe.

Returns:

Pandas DataFrame with id and various disabilities/impairments columns coded as 0 or 1.

Return type:

Pandas DataFrame

comorbidipy.calculator.hfrs(df: DataFrame, id: str = 'id', code: str = 'code')[source]

Calculate Hospital Frailty Risk Score

This is only applicable to patients who are 75 years or older.

https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(18)30668-8

Parameters:

  • df (pd.DataFrame) – Dataframe with 2 columns named id and code

  • id (str, optional) – Name of column to use as id. Defaults to “id”.

  • code (str, optional) – Name of column to use as code. Defaults to “code”.

Returns:

Dataframe with id and hfrs values.

Return type:

pd.DataFrame