In our study, a higher proportion of patients with diabetes had longer pre hospital delay than patients without diabetes when suffering a first myocardial infarction. Among patients with diabetes there were no differences in pre-hospital delay time between men and women however, the largest risk difference for pre-hospital delay ≥ 2 h was between women with and without diabetes. Diabetes, older age and living in towns or rural areas were all factors associated with pre-hospital delay ≥ 2 h. Our study is unique because it includes solely patients with first myocardial infarction, and it describes pre-hospital delay among both men and women with and without diabetes.
Our results are similar to previous research, which has also found that patients with diabetes have longer pre-hospital delay than patients without diabetes
[7, 11, 16, 17]. In contrast, some studies have found no differences in pre-hospital delay between patients with and without diabetes
[19, 27, 28]. However, those studies had fewer patients (n=140-403)
[19, 27], included patients with acute coronary syndrome
, and included both first and recurrent MI
[19, 27, 28].
In a recent report from the Northern Sweden MONICA Study, and in our present study, typical MI symptoms were common and atypical symptoms were not more frequent among patients with diabetes than in patients without diabetes
. Differences in typical/atypical symptoms between patients with and without diabetes could thus not be the explanation for the difference in pre-hospital delay between the groups. A possible explanation could be that patients with diabetes, like other patients with chronic illnesses, adjust to their symptoms and therefore ignore or overlook new symptoms
[29, 30]. In addition, the interpretation of MI symptoms as cardiac may be masked by symptoms associated with diabetes
[30, 31]. Vague MI symptoms such as dizziness and sweating can be interpreted as symptoms of low blood sugar level and not as symptoms of MI
In addition to diabetes, older age and living in towns or rural areas were also factors associated with pre-hospital delay ≥ 2 h. These results are consistent with prior studies, which have also found that older age is associated with longer pre-hospital delay
[7, 11, 16, 17]. One study investigated residential area as a factor associated with pre-hospital delay. In contrast to our results, no association was found between size of residential area and pre-hospital delay > 160 min
. In our study, size of residential area indirectly reflects distance to hospital but probably also socio-economic differences which may partly explain differences in delay.
Presenting with atypical MI symptoms was not in our result a predictor for pre-hospital delay ≥ 2 h. This is in contrast to most previous studies, which report that typical symptoms, such as chest pain, and interpreting symptoms as cardiac in origin predict shorter pre-hospital delay
. Similar to our results, two studies have reported that patients with atypical symptoms had shorter pre-hospital delay
 and no association was found between chest pain and delay
. Atypical symptoms such as syncope and dyspnea could be perceived with high intensity and as threatening. Fear, and experiencing symptoms as serious and/or life-threatening, have been described as factors for shorter pre-hospital delay times
[33, 34]. It is also possible that emotional factors such as anxiety can play an important role in the decision making process
As stated earlier, shortening the time interval between symptom onset and reperfusion is crucial for reducing mortality in MI
, and patients with diabetes have higher mortality from MI than patients without diabetes
[2, 3]. The longer pre-hospital delay among patients with diabetes may contribute to their higher mortality in MI compared with non-diabetics. Therefore, it is of utmost importance for health care personnel to educate patients with diabetes and to inform them about how to respond to symptoms of MI in order to shorten pre hospital delay.
The major strengths of our study are the large sample size (n=4266) and the fact that the large population-based database is controlled internally and externally for quality. Furthermore, it is a strength that all MI events are registered in the MONICA infarction registry and not only those treated in cardiology departments. To only include first MI strengthens the specificity of the findings and removes possible learning effects from a previous experience of seeking care for MI. Information about time between symptom onset and medical presence was missing in the medical records in 18% of the observations, and were thus excluded from the analysis. However; there were no major differences regarding diabetes status, gender, age and size of residential area between the excluded observations and the study population.
A limitation, however, is that information about pre-hospital delay time, in the MONICA registry was based solely on medical records. It has been discussed that other pre-hospital delay times are documented in medical records than in patient interviews
. Medical records are often made directly at admission to hospital, which can make it easier for the patient to remember times. Patient interviews are often made some days after arrival to hospital and therefore a risk for recall bias. It can also be difficult, both for patients and health care personnel, to delineate symptom onset of MI from prodromal symptoms, i.e. pre infarction angina before the acute MI event
A further limitation is the lack of data on socio-economical status in the MONICA Study. As diabetes is more common in people with low education and both these factors are more common in rural areas, the relationships noted could be confounded. In a sensitivity analysis we studied the differences in delay time restricted to urban dwellers and found similar longer delay in diabetic subjects as for the whole group (data not shown).