The main findings of the current study can be summarised as follows. First, following both STEMI and NSTEMI, SES1 patients were less frequently treated by PCI and more frequently by CABG as compared to SES4 patients. Second, the use of OMT is moderate in both SES1 and SES4 patients. Following STEMI, SES1 patients more frequently used complete OMT as compared to SES4 patients. Following NSTEMI, there was no difference in the rate of complete OMT use. Lastly, combining claims data with area-specific socioeconomic statistics is an efficient method to analyse cardiac care on a locoregional level in a unique way.

Our study shows that low SES STEMI and NSTEMI patients are treated by CABG more frequently than high SES patients. High CABG frequencies among low SES patients could be the result of more complex coronary lesions in this group. Previous studies have shown that, overall, low SES patients exhibit more risk factors (smoking, hypertension, hypercholesterolaemia and diabetes) at presentation, resulting in multivessel disease rather than one-vessel disease [15,16,17]. The relatively high frequency of CABG procedures in the low SES patients hints at the possibility of this population being unhealthier. Importantly, the findings of the current study provide a rationale for a conjoined initiative by cardiologists, family physicians and healthcare insurance companies to improve lifestyle in low SES regions.

OMT use was modest in STEMI and NSTEMI patients, among both low and high SES classes. This finding is not in line with a more recent study by Lee et al., which addressed the use of antihypertensive medication in event-free patients from different socioeconomic backgrounds, showing that low SES patients were less reluctant to adhere to medication [18]. However, an important difference from this study is that the event-free population is not comparable to our cardiovascular-event population. Furthermore, the presented study is in line with previous Dutch [6, 8, 19], British and American studies [20,21,22], all stressing the need for increased awareness of medication adherence after myocardial infarction. When addressing individual medication use after STEMI, foremost aspirin, beta blockers and statins are used more frequently by low SES patients than by high SES STEMI patients. In NSTEMI patients, low and high SES patients show a comparable low-usage pattern. Although some previous studies show that low SES has a negative effect on medication adherence [18, 23, 24], others have shown a negative effect of high SES [25, 26] or no effect of SES on medication adherence [27]. The reason for the observed difference in our study is difficult to ascertain, as no clinical data such as allergies or side-effect patterns were used. Lower use of cardioprotective medication in high SES STEMI patients can be related to an unwillingness to take medication because of doubts or a fear of side-effects, as observed in more highly educated patients [21, 28]. It could, however, also be related to a healthier lifestyle and less prevalent risk factors. Equally, a higher adherence among low SES STEMI patients could result from more CABG procedures being performed, having an effect on medication adherence in younger, low SES patients. As medication adherence is indispensable for survival after myocardial infarction, initiatives focusing on medication adherence in the high SES patients in the outpatient follow-up by cardiologists and primary care physicians are warranted.

For the first time in Dutch healthcare, claims data were combined with area-specific socioeconomic statistics. These data illustrate that treatment patterns and healthcare use in specific regions and specific patient groups can be analysed by this approach. This type of research differs from previously performed ‘causality studies’ assessing the impact of low SES on mortality or adverse events after myocardial infarction [16, 28, 29].

Our findings stress the importance of primary prevention programmes for myocardial infarction patients in low SES regions and equally provide a rationale for the development of quality improvement programmes which focus on medication adherence after myocardial infarction among inhabitants of high SES regions. By aiming at inhabitants of specific regions, scarce healthcare resources are spent more effectively and care becomes more patient-centred. For example, medication use after myocardial infarction could be the topic of patient-education gatherings in general practices or cardiac outpatient clinics to increase awareness. Furthermore, understanding the patient’s experience regarding medication use has to be addressed further on a qualitative level via, for example, questionnaires and interviews, as a local study has shown [30].

Some limitations should be considered when interpreting the results. First, the study uses observational data. Second, the SES score used is a postal-code average and not derived on a personal level: some SES4 patients might live in an SES1 area and vice versa. Additionally, yearly income or level of education were not separately collected per patient in our study. Third, the level of clinical details per patient is limited. Accordingly, completeness of revascularisation, ventricular function or infarct size are not included and commonly used risk scores (e.g. GRACE risk score) cannot be applied to our study population. Lastly, the total number of CABG procedures performed in our population could be underestimated because one off-site PCI centre was included.