Use of administrative data to analyse comorbidities in hospitalized patients for ischemic stroke in Wallonia

Abstract

Ischemic stroke is a major cause of morbidity and mortality in European countries. The risk factors for stroke are either modifiable as high blood pressure, high cholesterol, diabetes, overweight or obesity, smoking or sedentary lifestyle or non-modifiable as age, sex, ethnicity and genetic factors. The importance of each risk factor may vary by age and sex. Knowledge of these variations may be important in terms of management of the acute event but also for prevention policies. The aim of this study is to analyse the comorbidities of acute ischemic strokes using routine administrative data. The present study was conducted in Wallonia from 1 January 2013 to 31 December 2014. Wallonia is the southern region of Belgium, with 3,563,060 inhabitants in 2013. The analyses presented below relates to patients aged 25 and over. The hospital admissions for stoke ischemic (ICD-9 code 433 to 435) were taken from the ‘Résumé Hospitalier Minimum’ (RHM). The RHM is a mandatory register kept by Belgian hospitals containing patient data (e.g.: year of birth, gender, place of residence), stay data (e.g.: admission date) and admission data (e.g.: principal and secondary diagnosis). Secondary diagnoses were considered as risk factors. Factorial correspondence analysis with graphical output was used to highlight risk factors across age and sex. This method is useful in analysing multivariate data to discover associations when a contingency table is not adequate. Correspondence analysis decomposes the chi-squared statistic associated with contingency table into orthogonal factors. A total of 14,868 hospitalisations for stroke were analysed. The median age at admission is 72 for men and 79 for women. Men represented 48% of stroke hospitalisations. A significant number of stroke patients presented with comorbidity in the form of a history of hypertension (55%), hypercholesterolemia (46.5%), cardiac arrhythmias (32.9%), cardiac ischemic diseases (26.9%), diabetes (24.0%), smoking (20.6%) and overweight (16.9%). Factorial correspondence analysis shows a large variation in age-related comorbidities profiles. Overweight is more pronounced between 55–64-years-old. Hypertension, hypercholesterolemia and diabetes are very close to the 65–84 age group. Heart problems (cardiac ischemic diseases and cardiac arrhythmias) are more prevalent for patients 85 and over. Finally, smoking and excessive alcohol consumptions are mostly present in the younger age groups. The comorbidity profile is not influenced by gender. The medico-administrative data of the RHM allowed to highlight profiles of comorbidities of acute ischemic stroke according to age and sex. The results obtained are consistent with risk factors for stroke described in the scientific literature. However, the present study is the first to explore profiles of comorbidities in Wallonia, a region in the south of Belgium. Factorial correspondence analysis is often used in the social sciences but rarely in the field of health. Our study is the first to use this statistical method to highlight patterns of comorbidities by age and sex of patients with acute stroke. This study suggests the possibility of using administrative data for epidemiological purposes, which could avoid planning costly data collection.