One of the aims of this study is to identify the determinants of pre-hospital delay for MI patients in the Northern region of Bangladesh. By knowing the factors that may influence the time spent before hospital admission after onset of MI, we may help to formulate strategies to reduce the delay in initiation of effective life-saving treatment for this condition.
The median pre-hospital delay of our study population was 9 (IQR 13) hours, while the median decision time was 2 (IQR 2.8) hours. Referral time from private hospitals was noted to be longer than that of the government hospitals. Overall, only 39.5% of our patients were admitted to RMCH within 6 hours of their symptom onset. Patients who had typical clinical presentation, suspected the symptoms as cardiac in origin, visited qualified doctors, and seek treatment directly with RMCH were more likely to be admitted within 6 hours after onset of symptom.
A study conducted in a Dhaka, the capital city of Bangladesh, reported that the mean pre-hospital delay for their MI patients was 11.67 hours, and about 77% of the them reached the hospital within 6 hours (15). That study was conducted in private tertiary care hospital, where most of patients were from affluent families. Patients from high income families in our study also showed shorter pre-hospital delay compared to others from lower- and middle-income families. Another study conducted in a government tertiary care hospital situated in Chittagong in Southern part of Bangladesh (16) reported early presentation (within 6 hours) in 17.2% of their patients. In Bangladesh, the set up and facilities of all government tertiary care hospitals were quite similar all over the country. We could expect that the main factors that contributed towards the discrepancy would be on the social demographic features and health care seeking behavior between these regions.
In developed countries, national level pre-hospital delay of MI patients were generally low. The Global Registry of Acute Coronary Events (GRACE) study reported more than 70% of patients presented to hospital within 6 hours, with a median of 3 hours (18). The pre-hospital delays reported by studies conducted in our neighboring developing countries like India were between 3.0 to 5.2 hours, and almost 60% of patients were admitted to hospital within 6 hours (19,20). Similar studies in Pakistan reported between 66 to 73% of patients presented to hospital within 6 hours (17,21).
In our study cohort, pre-hospital delay for both young and elderly patients were more likely to be more than 6 hours. The conventional belief that MI is the disease of the old age influenced many younger patients to attribute some cardiac symptoms to less critical conditions like heartburn or peptic ulcer disease, thus hindered early intervention. Older age had also been noted to be risk factor of pre-hospital delay (16,18,22) , and this may be due to limited resources and problem with transportation. There was no significant difference in pre-hospital delay between both the genders in our study, and this was consistent with most other studies from both developed and developing countries (10,16,23–25). However, some studies identified female gender as one of the predictors of pre-hospital delay (12, 26).
Patients from rural areas and lower income families were more vulnerable to delayed in hospital admission, though it might not be related with their educational qualification. Previous studies have reported that residence from rural areas (23,27) and those from lower socio-economic backgrounds (28) were at higher risk of delayed hospital admission due to lack of financial resource and availability of transportation. In addition, those staying closer than 5.0 km from primary care facilities were more likely to reach the hospital within 6 hours after onset of MI symptoms. Similar finding had been reported where patients attributed long distance from primary care facilities as the main reason for their late presentations (23,25,29). However, our study showed that the mode of transportation (ambulance or general vehicle) was not associated pre-hospital delay. Interesting, a study from South India reported that patients who used private ambulance had shorter delay compared to those who used public ambulance or other types of transportation (27).
We were not able to show any association between status of education and pre-hospital delay. There were conflicting information in the literature, where some studies reported higher educational qualification reduces the pre-hospital delay (12,23,28), while others indicated no relationship between the two (27,30). We observed an interesting finding where patients in the middle-income group had a much higher odd for delay in hospital admission than those from the lower-income group. We were not able to explain this findings, and further study into this phenomenon is probably indicated.
We noted that clinical symptoms at the onset of MI was a significant predictor for delay in hospital admission. Patients who presented with chest pain were more likely to be admitted for treatment within 6 hours of onset of MI, compared to those with other pain over other sites, or atypical and vague symptoms, and this was also reported by other studies (18,28). Patients tend to misinterpret the atypical symptoms to be of non-cardiac in origin, and this would hindered them from seeking urgent medical attention. Although Non-ST-elevated Myocardial Infarction (N0n-STEMI) has been known to present with atypical symptoms, which can affect the pre-hospital delay of some patients, they did not influence the pre-hospital delay in our patients.
Patients’ behavior and primary action after onset of symptoms were important factors that closely associated with pre-hospital delay. Patients who suspected they had MI were more likely to have early hospital admission while those who considered the symptoms as nothing serious, or decided to wait for spontaneous resolution were more likely to have delayed hospital admission. This finding was similar to other studies where perceived susceptibility to MI was associated with shorter pre-hospital delay, while misinterpretation of symptoms or pain resistance behavior were associated with longer pre-hospital delay (31,32). Visiting non-qualified or under-qualified medical practitioners, consulting drug sellers or self-medication significantly increased pre-hospital delay. Though this practice is rare in developed countries, these behaviors markedly increased the pre-hospital delay (33, 34). Other studies have reported that even visiting general practitioners would increase pre-hospital delay (19,25).
Although the patients admitted directly to RMCH had a longer median decision to hospital arrival time, most of them were able to reach the hospital within 6 hours. This difference is obvious as most of the patients who have visited government primary health centers were from the rural areas whereas those who were admitted directly to RMCH were mainly from nearby urban areas. The finding was consistent with previous studies from different countries (11,25,35). We did noted another study from Bangladesh that reported the opposite finding, where patients who visited primary care centers were more likely to present early (16). However, their definition of early presentation was 12 hours after onset of symptoms. Our study noted that longer referral time for patients from private hospitals. It was possible that in private setting, there was a higher tendency for doctors to re-establish the diagnosis before they send the patients off to another institution.
Influence of long standing co-morbidities on pre-hospital delay remains uncertain. Our study shows that patients with diabetes mellitus or previous history of coronary arterial diseases were more likely to be admitted earlier compared to their counterparts, but no positive correlation was noted for those with hypertension. Some studies reported that patients with diabetes mellitus and previous history of cardiovascular disease were associated late presentation (18,36). On the other hand, our study showed that positive family history of cardiovascular disease was associated with early presentation to the hospital, and this was consistent with other studies (16,33). Perhaps in our population, patients and family members with chronic diseases were generally more health conscious and more familiar with available resources, and these would influence them to seek medical attention early.
In-hospital mortality was noted to be higher among the patients who were admitted after 6 hours, even after adjusting for other potential confounding factors like age, sex, type of MI and other comorbidities. Similar finding was reported by another study conducted on German population, where unknown or prolonged pre-hospital delay was associated with increased in-hospital death (37). Other studies has also shown that treatment efficacy of MI decreases with time after onset of symptoms, and this would increase the risk of mortality (2,6,7).
Our study provides a clear insight on various factors that were associated with pre-hospital delay of MI patients in Northern regions of Bangladesh. Patients’ behavior and health seeking actions would increase the pre-hospital delay. Mis-information and wrong care seeking behavior were most likely due to lack of knowledge and awareness about common symptoms of MI, (8). Public awareness about symptoms of MI should be raised so that misinterpretations can be reduced. Moreover, patients should be encouraged to visit qualified physicians or hospitals and medical practice of non-qualified personnel should be restricted to reduce wrong diagnoses that may proof to be fatal for MI patients. Referral time, especially from private primary care hospitals should be reduced. Providing emergency diagnostic support in the primary care centers like ECG and cardiac troponin tests can reduce the referral time and thereby reduce the total pre-hospital delay. Proper use of ambulance service and the newly launched emergency call number (999) may reduce the delay, and these observations have been reported in other countries like Sweden (25).
Strengths and limitations: One of the strengths of this study is that the sample population was from the Northern region of Bangladesh, and this covered a large portion of the national population. Moreover, patients’ symptoms, diagnoses and other clinical data were obtained from the medical records that reduced the risk of recall bias. The main limitation of this study is that it does not represent the overall population of the country. This was a cross sectional single-center study which doesn’t reflect cover the whole population sample of the country. The time of onset of symptoms and primary actions were based on patients’ statement gathered during the interview, and there was a possibility of recall bias. Moreover, convenience sampling method may also introduce sampling bias. This study did not include patients who did not reach the hospital, died shortly after admission, or died before they were fit for interview. Other than death during hospital stay, we did not study other forms of morbidities related to MI or its treatment.