Result tables description

Result of PheWAS analysis

Variable Name	Type	Description
phecode	String	Disease code (Phecode) used in PheWAS analysis
disease	String	Disease name corresponding to the Phecode
system	String	Phecode disease system corresponding to the Phecode (e.g., infectious diseases)
sex	String	Sex-specificity of the disease (e.g., Both, Male, Female)
N_cases_exposed	Integer	Number of individuals diagnosed with the disease in the exposed group
describe	String	Descriptions of the model fitting state and removed covariates with reasons
exposed_group	String	Incidence rate (unit: per 1,000 person-years) in the exposed group
unexposed_group	String	Incidence rate (unit: per 1,000 person-years) in the unexposed group
phewas_coef	Float	Estimated coefficient from the model
phewas_se	Float	Standard error of the estimated coefficient
phewas_p	Float	P-value indicating statistical significance of the coefficient
phewas_p_significance	Boolean	Indicates whether the result is statistically significant based on adjusted p-value (True/False)
phewas_p_adjusted	Float	Adjusted p-value accounting for multiple comparisons

Result of comorbidity strength estimation

Variable Name	Type	Description
phecode_d1	Integer	Phecode for disease 1 in the disease pair
phecode_d2	Integer	Phecode for disease 2 in the disease pair
name_disease_pair	String	Name of the disease pair (format: “D1-D2”)
N_exposed	Integer	Total number of individuals in exposed group
n_total	Integer	Number of exposed individuals included in the sub-cohort that meet the sex-specificity eligibility criteria for both diseases and after excluding those with history of either disease 1, disease 2, or related diseases
n_d1d2_diagnosis	Integer	Number of individuals diagnosed with both diseases
n_d1_diagnosis	Integer	Number of individuals diagnosed with disease 1
n_d2_diagnosis	Integer	Number of individuals diagnosed with disease 2
n_d1d2_nontemporal	Integer	Number of individuals diagnosed with both disease 1 and disease 2 but without defined temporal order (i.e., the time interval between the two diagnosis is smaller than or equal to min_interval_days or larger max_interval_days)
n_d1d2_temporal	Integer	Number of individuals diagnosed with disease 1 followed by disease 2 in a defined temporal order (i.e., the time interval between the two diagnosis is larger than min_interval_days and smaller than or equal tomax_interval_days)
n_d2d1_temporal	Integer	Number of individuals diagnosed with disease 2 followed by disease 1 in a defined temporal order (i.e., the time interval between the two diagnosis is larger than min_interval_days and smaller than or equal tomax_interval_days)
phi_coef	Float	Phi coefficient (φ), Pearson’s correlations for two binary variables
phi_p	Float	P-value for Phi coefficient significance
RR	Float	Relative risk of observing both conditions in the same individual relative to expectation
RR_p	Float	P-value for relative risk
phi_p_adjusted	Float	Adjusted P-value for Phi coefficient (multiple comparisons)
RR_p_adjusted	Float	Adjusted P-value for relative risk (multiple comparisons)
phi_p_significance	Boolean	Whether the Phi is statistically significant based on adjusted p-value
RR_p_significance	Boolean	Whether the RR is statistically significant based on adjusted p-value
disease_d1	String	Name of disease 1
system_d1	String	Phecode disease system related to disease 1
sex_d1	String	Sex-specificity of disease 1
disease_d2	String	Name of disease 2
system_d2	String	Phecode disease system related to disease 2
sex_d2	String	Sex-specificity of disease 2

Result of binomial test

Variable Name	Type	Description
phecode_d1	Float	Phecode for disease 1 in the temporal disease pair
phecode_d2	Float	Phecode for disease 2 in the temporal disease pair
name_disease_pair	String	Name of the temporal disease pair (e.g., D1->D2)
n_d1d2_nontemporal	Float	Number of individuals diagnosed with both disease 1 and disease 2 but without defined temporal order (i.e., the time interval between the two diagnosis is smaller than or equal to min_interval_days or larger max_interval_days)
n_d1d2_temporal	Float	Number of individuals diagnosed with disease 1 followed by disease 2 in a defined temporal order (i.e., the time interval between the two diagnosis is larger than min_interval_days and smaller than or equal tomax_interval_days)
n_d2d1_temporal	Float	Number of individuals diagnosed with disease 2 followed by disease 1 in a defined temporal order (i.e., the time interval between the two diagnosis is larger than min_interval_days and smaller than or equal tomax_interval_days)
binomial_p	Float	P-value from the binomial test for directionality
binomial_proportion	Float	Proportion of successful outcomes in the binomial test
binomial_proportion_ci	String	Confidence interval for the binomial proportion
disease_d1	String	Name of disease 1
system_d1	String	Phecode disease system for disease 1
sex_d1	String	Sex-specificity of disease 1
disease_d2	String	Name of disease 2
system_d2	String	Phecode disease system for disease 2
sex_d2	String	Sex-specificity of disease 2
binomial_p_significance	Boolean	Indicates whether the result is statistically significant based on adjusted p-value
binomial_p_adjusted	Float	Adjusted p-value for multiple comparisons

Result of comorbidity network analysis

Variable Name	Type	Description
phecode_d1	Float	Phecode for disease 1 in the non-temporal disease pair
phecode_d2	Float	Phecode for disease 2 in the non-temporal disease pair
name_disease_pair	String	Name of the non-temporal disease pair (e.g., “D1-D2”)
N_exposed	Integer	Total number of individuals in exposed group
n_total	Integer	Number of exposed individuals included in the sub-cohort that meet the sex-specificity eligibility criteria for both diseases and after excluding those with history of either disease 1, disease 2, or related diseases
n_exposed/n_cases	String	Number of exposed individuals (individuals with diagnosis of D1) among cases (individuals with diagnosis of D2)
n_exposed/n_controls	String	Number of exposed individuals (individuals with diagnosis of D1) among controls (individuals without diagnosis of D2)
comorbidity_network_method	String	Method used for comorbidity network analysis
describe	String	Description of the model fitting, removed covariates in the model, and reasons for removal of covariates in the model
co_vars_list	String	List of covariates used in the model
co_vars_zvalues	String	Z-values for each covariate in the model
comorbidity_beta	Float	Estimated coefficient from the comorbidity model
comorbidity_se	Float	Standard error of the estimated coefficient
comorbidity_p	Float	P-value for the comorbidity coefficient
comorbidity_aic	Float	Akaike information criterion for the model
disease_d1	String	Name of the disease 1
system_d1	String	Phecode disease system for the disease 1
sex_d1	String	Sex-specificity of the disease 1
disease_d2	String	Name of the disease 2
system_d2	String	Phecode disease system for the disease 2
sex_d2	String	Sex-specificity of the disease 2
comorbidity_p_significance	Boolean	Whether the result is statistically significant based on adjusted p-value
comorbidity_p_adjusted	Float	Adjusted p-value accounting for multiple comparisons
Columns for RPCN method
alpha	Float	Hyperparameter used for l1-norm (Weight multiplying the l1 penalty term)
Columns for PCN_PCA method
pc_sum_variance_explained	Float	The cumulative proportion of variance that is accounted for by a selected number of principal components in a Principal Component Analysis (sum of explained variance for principal components)

Result of disease trajectory analysis

Variable Name	Type	Description
phecode_d1	Float	Phecode for disease 1 in the temporal disease pair
phecode_d2	Float	Phecode for disease 2 in the temporal disease pair
name_disease_pair	String	Name of the temporal disease pair (e.g., “D1 → D2”)
N_exposed	Integer	Total number of individuals in exposed group
n_total	Integer	Number of exposed individuals included in the nested case-control dataset, where eligible cases (with diagnosis of D2) are all selected and matched with specified number of controls using incidence density sampling
n_exposed/n_cases	String	Number of exposed individuals (individuals with diagnosis of D1) among cases (individuals with diagnosis of D2)
n_exposed/n_controls	String	Number of exposed individuals (individuals with diagnosis of D1) among cases (individuals with diagnosis of D2)
trajectory_method	String	Method used for disease trajectory analysis
describe	String	Description of the model fitting, removed covariates in the model, and reasons for removal of covariates in the model
co_vars_list	String	List of covariates included in the model
co_vars_zvalues	String	Z-values for each covariate in the model
trajectory_beta	Float	Estimated coefficient from the model
trajectory_se	Float	Standard error of the estimated coefficient
trajectory_p	Float	P-value for the coefficient
trajectory_aic	Float	Akaike information criterion for the model
disease_d1	String	Name of the disease 1
system_d1	String	Phecode disease system for the disease 1
sex_d1	String	Sex-specificity of the disease 1
disease_d2	String	Name of the disease 2
system_d2	String	Phecode disease system for the disease 2
sex_d2	String	Sex-specificity of the disease 2
trajectory_p_significance	Boolean	Whether the result is statistically significant based on adjusted p-value
trajectory_p_adjusted	Float	Adjusted p-value accounting for multiple comparisons
Columns for RPCN method
alpha	Float	Hyperparameter used for l1-norm (weight multiplying the l1 penalty term)
Columns for PCN_PCA method
pc_sum_variance_explained	Float	The cumulative proportion of variance in a dataset that is accounted for by a selected number of principal components in a Principal Component Analysis (sum of explained variance for principal components)