Article type
Year
Abstract
Background: Quality information is useful for monitoring population health-status, and planning and evaluating health care. Morbidity indicators, based on aggregated data from health information systems (HIS), provide information about disease prevalence.
Objectives: To determine the prevalence of obesity and diabetes mellitus and to assess data quality for diagnosis coding, among registered patients over 19 years old for obesity, and over 44 years old for diabetes, in health centers of the Portuguese central health region, from December 2013 to January 2014. Methods: Prevalence of the disease was calculated as the proportion of patients registered with a diagnostic code, according to the International Classification of Primary Care (ICPC-2). The proportion of patients with an unknown disease status, due to out of date information or missing data, was computed. Data quality was assessed using quantitative attributes, defined by the Centers for Disease Control and Prevention to evaluate public health surveillance systems. Compliance of the clinical criteria, defined in national guidance issued by the Portuguese Directorate-General of Health or other scientific societies, with an ICPC-2 registered diagnostic code, was measured. Results: Prevalence of obesity among 1,087,629 registered patients was 6%; prevalence of diabetes was 14% (n = 659,914). Disease status was unknown in 31% to 60% of patients. Representativeness of the registered patients with medical appointments was over 90%. HIS completeness for body mass index registry was 45% (n = 1,001,847) while for fasting plasma glucose/haemoglobin A1c and/or insulin/oral anti-diabetic drugs it was 70% (n = 602,903). Sensitivity for ICPC-2 coding among patients with clinical criteria was 43% (n = 344,712) for obesity, and 90% (n = 399,063) for diabetes. Conclusions: Considering representativeness, the proportion of patients with an unknown disease status was unexpectedly high. Higher completeness and sensitivity showed higher prevalence of the disease, highlighting differences in prevalence, according to differences in data quality. Suggested data validation rules could lead to improved quality.
Objectives: To determine the prevalence of obesity and diabetes mellitus and to assess data quality for diagnosis coding, among registered patients over 19 years old for obesity, and over 44 years old for diabetes, in health centers of the Portuguese central health region, from December 2013 to January 2014. Methods: Prevalence of the disease was calculated as the proportion of patients registered with a diagnostic code, according to the International Classification of Primary Care (ICPC-2). The proportion of patients with an unknown disease status, due to out of date information or missing data, was computed. Data quality was assessed using quantitative attributes, defined by the Centers for Disease Control and Prevention to evaluate public health surveillance systems. Compliance of the clinical criteria, defined in national guidance issued by the Portuguese Directorate-General of Health or other scientific societies, with an ICPC-2 registered diagnostic code, was measured. Results: Prevalence of obesity among 1,087,629 registered patients was 6%; prevalence of diabetes was 14% (n = 659,914). Disease status was unknown in 31% to 60% of patients. Representativeness of the registered patients with medical appointments was over 90%. HIS completeness for body mass index registry was 45% (n = 1,001,847) while for fasting plasma glucose/haemoglobin A1c and/or insulin/oral anti-diabetic drugs it was 70% (n = 602,903). Sensitivity for ICPC-2 coding among patients with clinical criteria was 43% (n = 344,712) for obesity, and 90% (n = 399,063) for diabetes. Conclusions: Considering representativeness, the proportion of patients with an unknown disease status was unexpectedly high. Higher completeness and sensitivity showed higher prevalence of the disease, highlighting differences in prevalence, according to differences in data quality. Suggested data validation rules could lead to improved quality.