Agreement in race-ethnicity coding between a hospital discharge database and another database.

Area of Science:

• Health Informatics
• Public Health
• Epidemiology

Background:

• Accurate race and ethnicity data are crucial for understanding health disparities.
• Hospital discharge databases are primary sources for health statistics but may have coding inconsistencies.

Purpose of the Study:

• To evaluate the agreement of race and Hispanic ethnicity coding between a hospital discharge database and a linked cancer registry.
• To identify potential improvements for race-ethnicity data ascertainment in hospital records.

Main Methods:

• Compared race-ethnicity codes for 72,276 cancer patients using linked hospital discharge data and a statewide cancer registry.
• Utilized patient birthplace and surname information for enhanced ethnicity ascertainment, including linkage to a Spanish surname list.
• Employed Kappa coefficients to quantify agreement between the two databases for various racial and ethnic categories.

Main Results:

• Substantial agreement was found for White (kappa=.74), Black (kappa=.93), Hispanic ethnicity (kappa=.73), and non-Hispanic White (kappa=.83) categories.
• Moderate agreement was observed for the Asian-Pacific Islander race category (kappa=.52).
• Significant discrepancies were noted, with only 42% of Asian-Pacific Islanders and 62% of Hispanics in the registry having matching codes in the discharge database.

Conclusions:

• Existing hospital discharge databases demonstrate variable accuracy in coding racial and ethnic groups.
• Incorporating birthplace and surname data into hospital discharge databases could significantly improve the ascertainment of specific racial-ethnic populations.
• Enhancing data collection methods is essential for reducing health disparities and improving public health surveillance.