Introduction

Ischemic heart disease (IHD) is an acute condition that involves a high risk of death and entails negative social and economic impacts. In 2016, IHD caused 17.3% of worldwide mortality, 18.2% in high sociodemographic index (SDI) countries, and 14.3% in low-and-middle-SDI countries.¹ Acute myocardial infarction (AMI) is the most devastating IHD.²

Mexico has higher AMI mortality than European and other Latin American countries. In 2015, the thirty-day AMI mortality after hospital admission among member countries of the Organization for Economic Cooperation and Development ranged between 3.7% in Norway and 28.1% in Mexico.³ Moreover, in-hospital mortality in Mexico (26.6%) is higher than in Nicaragua (25%), Peru (12%), Argentina (8.8%), Brazil (8.8%), and Colombia (6%).^4,5,6,7,8

Health outcomes of AMI treatment depend on factors related to both the patient and the health services. On the patient’s side, women, older adults, low socioeconomic status individuals, and those uninformed regarding suspicion of AMI symptoms have more probabilities of delaying health-seeking care, or of not receiving specialized care, and, therefore, dying.^9,10,11 While on the health services side, there are pre-hospital and in-hospital factors that influence health outcomes. Pre-hospital factors include shortage of ambulances equipped with electrocardiography and capability to transmit electrocardiograms (EKG) to the on-call cardiologist.^12,13,14 In-hospital factors include scarcity of medical doctors trained to diagnose and treat AMI, absence of reperfusion therapy, inadequate adherence to reperfusion therapy guidelines, shortage of thrombolytic drugs, reduced availability of 24-hour hemodynamic rooms, and limited skills of cardiologists to perform percutaneous coronary interventions (PCI).^15,16 Favorable health outcomes of AMI can be achieved when the supply side (i.e. cardiologists, equipment, medicines) meets the patients’ needs for access to, and availability and high quality of care.

The pervasive high AMI mortality rates in Mexico justify an in-depth analysis of the supply and demand of AMI health care. On the demand side, estimating the trends of AMI hospitalizations allows ascertaining the changes in its magnitude. On the supply side, an analysis of the competence of the Ministry of Health (MOH) hospitals to treat AMI may inform about existing gaps. Moreover, estimating the in-hospital mortality, which is a short-term health outcome, signals the clinical performance to manage the acute phase of AMI.

Therefore, the objectives of the present study were to analyze the trends of AMI hospitalizations and in-hospital mortality rates between 2005 and 2017 and evaluate the competence of MOH hospitals to diagnose and treat this condition.

Materials and methods

We used a two-stage mixed methods approach. The first stage analyzed the trends of AMI admissions and in-hospital mortality rates between 2005 and 2017 of all MOH hospitals. The second stage consisted in a nation-wide cross-sectional online survey to ascertain the competence of MOH hospitals for providing AMI care.

Stage 1: Analysis of trends in hospitalizations and in-hospital mortality. We analyzed the data of hospital discharges and deaths registered in the MOH Health Information System from January 1st, 2005 to December 31st, 2017. The analysis considered the nearest year when the states did not have information for 2005. We selected the registries with diagnoses of AMI according to the following codes of the 10th revision of the International Classification of Diseases (ICD-10): I210-I214, I219, I220, I221, I228, I229, I230- I235.

The AMI hospitalization rate formula was the number of AMI hospitalizations in MOH facilities for each age-sex group, divided by the total number of people (stratified by age and sex) covered by the MOH, and multiplied by 100 000.

The estimation of the AMI hospitalization rates at national and state levels included patients aged 30-60 years and >60 years. We stratified these age groups because the extinct public healthcare insurance program Seguro Popular only provided funds for the treatment of AMI patients aged <60 years until 2017. Patients aged >60 years did not receive this benefit; MOH hospitals treated them with their regular resources. The annual reports of the former Seguro Popular and the national census were the sources for estimating the amount of people that the MOH covered. The AMI in-hospital mortality rate formula was the number of AMI patients that died during their hospital stay for each age-sex group, divided by the total number of hospitalized AMI patients of each age-sex group and multiplied by 100.

Stage 2: From May to August 2016 we conducted an online cross-sectional survey to ascertain the hospitals’ competence for providing AMI diagnosis and treatment. The survey sample included all secondary and tertiary level MOH hospitals in the 32 states (n=893); we excluded pediatric hospitals. The General Directorate for Quality of Healthcare and Education (DGCES - acronym in Spanish) of the MOH led the survey. Subsequently, the state quality managers of the DGCES followed up by requesting the local health authorities and hospitals to answer the online questionnaire.

A group of decision-makers (DGCES), cardiologists from the National Institute of Cardiology and researchers from the Mexican Institute of Social Security, and the Inter-American Development Bank (IDB) designed and tested the online questionnaire in two hospitals before distributing it to participating hospitals.

Hospital competence was defined as follows: 1. Hospital competence for diagnosing AMI: availability of personnel trained to diagnose AMI, electrocardiogram equipment (EKG), and enzymes test. 2. Hospital competence to perform pharmacological reperfusion: staff trained, availability of electrocardiograph (EKG), crash cart, antithrombotic treatment, protocol for pharmacological reperfusion and for referring patients to facilities with hemodynamic room. 3. Hospital competence to perform mechanical reperfusion: Personnel trained to perform mechanical reperfusion, availability of EKG, crash cart, antithrombotic treatment, pharmacological reperfusion, and 24/7 hemodynamic room.

In order to identify and measure the hospitals’ competence for providing AMI healthcare, the questionnaire covered: 1. Hospital level of care, according to the classification of the MOH, we identified secondary and tertiary care hospitals; infrastructure: laboratory availability of cardiac enzymes tests; equipment such as crash cart, functional defibrillators, electrocardiographs, hemodynamic rooms and medications such as thrombolytic medications, aspirin, heparin, and clopidogrel. 2. Healthcare workforce availability such as a cardiologist and staff trained to perform AMI diagnosis, thrombolytic therapy, and coronary angioplasty.

Results

Stage 1. Results of the trend analysis for hospitalizations and in-hospital mortality due to AMI during 2005-2017.

AMI hospitalization rates were higher in men compared to women and in patients aged >60 years compared to those 30-60 years old. Figure 1 depicts the trends from 2005 to 2017 of AMI hospitalization rates (per 100 000 population) by sex and age at a national level. In men aged 30-60 years, the AMI hospitalization rate increased from 8.9 to 13.1. In men aged >60 years, it increased from 40.4 to 58.4. The AMI hospitalization rate per 100 000 increased from 2.7 to 4.1 in women aged 30-60 years, and from 25.0 to 32.3 in women aged >60 years.

* p-value of <0.05

Figure 1 Acute myocardial infarction hospitalization rates between 2005-2017. Mexico 

The AMI in-hospital mortality was higher in women compared to men, and in people older than 60 years compared to those aged 30-60 years. Figure 2 shows national trends in AMI in-hospital mortality rates (per 100 discharges). The AMI in-hospital mortality rate showed a slight upward trend from 2005 to 2017; although it was statistically significant in men aged 30-60 years, during 2005-2017, it increased from 11.5 to 14.6 deaths per 100 discharges (APC 1.8; 95%CI: 0.1, 3.5).

* p-value of <0.05

Figure 2 Acute myocardial infarction in-hospital mortality rates between 2005-2017. Mexico 

The analysis of hospitalization rates per 100 000 adults showed an upward tendency (table I). The increase in hospitalization rates was statistically significant in 19 states. The highest increases among men aged 30-60 years happened in Aguascalientes, Sinaloa, Sonora, and Zacatecas, and in Guanajuato, Sonora and San Luis Potosi, among men aged >60 years. Among women aged 30-60 years it rose in eight states, mainly in Michoacán, Guanajuato, and Puebla, and among women >60 years old, in Sinaloa, Guanajuato, and Tabasco, and decreased in Colima and Tamaulipas.

Between 2005 and 2017, the in-hospital mortality rate exhibited an upward tendency among men aged 30-60 years in Chihuahua, Jalisco, and Veracruz, and in men aged >60 years, in Chihuahua, Guanajuato and Tamaulipas. It decreased in Aguascalientes, Campeche, and Puebla (table II). Among women aged 30-60 years, it increased in Hidalgo and Veracruz, and in Guanajuato and Morelos among women aged >60 years, and decreased in Aguascalientes and Nayarit. The highest rates of in-hospital mortality happened in women aged >60 years, in comparison to men of the same age group.

Stage 2. The supply capacity of MOH hospitals for the diagnosis and treatment of AMI.

The survey included 527 MOH hospitals (61% of 893 MOH hospitals), of which 471 (89%) were secondary care level hospitals, and 56 (11%) were tertiary care level hospitals. The non-response rate was 39%. We estimated that 26.3 million people over 30 years of age were eligible to receive care in these facilities.

We analyzed the competency of hospitals to diagnose AMI and to perform pharmacological and mechanical reperfusion (table III). Regarding the secondary level hospitals, only 16% were certified by the General Health Council (GHC). Overall, 37% had the competency to diagnose AMI; only 8.7%, to perform pharmacological reperfusion, and 2.8%, to perform mechanical reperfusion. As for tertiary level hospitals, 12% were GHC certified, 51% were competent to diagnose AMI; 26%, to perform pharmacological reperfusion, and 18%, to carry out mechanical reperfusion. There were wide gaps in hospital capacity between states. We detected that nine of the secondary level hospitals (Baja California Sur, Colima, Durango, Hidalgo, Nayarit, Tabasco, Tlaxcala, Yucatán, and Zacatecas) did not have the competence to perform pharmacological reperfusion, while secondary level hospitals in 18 states were not competent to perform mechanical reperfusion. As for tertiary level hospitals, only 12% were CGE certified, and the hospitals of the states of Mexico, Guerrero, Puebla, Sinaloa, Tlaxcala, Veracruz, Yucatán, and Zacatecas had no competence to perform pharmacological or mechanical reperfusion.

Between 2010 and 2015, the hospitals treated 40 071 AMI patients; only 27% received thrombolysis, and 12% underwent angioplasty. The percentage of patients that underwent reperfusion varied by state, ranging from zero reperfusion in Colima, Michoacán, Quintana Roo hospitals, to 54.2% of patients treated with pharmacological reperfusion in Mexico City, and 65.2% treated with mechanical reperfusion in Puebla.

Discussion

The main results of the study indicate an upward trend of AMI hospitalizations in men aged 30-60 and >60 years, and in women aged >60 years. Besides, in-hospital mortality rates in men and women >60 years of age are on the rise, and MOH hospitals have little competence to diagnose AMI, exhibiting wide disparities, or to perform pharmacological and mechanical reperfusion.

The increasing trend in AMI hospitalizations among patients aged >60 years is similar to what low-and-middle-SDI countries report.¹⁹ The finding is opposite to the downward trends in AMI hospitalizations in high-SDI countries,^20,21,22 where successful preventive programs encourage the use of outpatient cardioprotective medications.²³ Effective prevention programs to reduce cardiovascular risk factors are an unmet need in Mexico.

In-hospital mortality is a health outcome and a proxy indicator of high quality care. It showed high rates mainly among men and women >60 years, signaling age and gender gaps. The wide disparities of in-hospital mortality rates might be related to the competence of the hospital to manage AMI. Subnational variations are not uncommon; they have been reported in China and Brazil.^24,25 Ascertaining subnational differences can guide targeted interventions to bridge in-hospital mortality gaps. Moreover, a study from Brazil reported that hospital mortality rates increased with age after adjusting for baseline risk differences.²⁶ Elderly patients have more complex comorbidities and worse outcomes and are less likely to undergo revascularization or receive acute and long-term medications.²⁷ Access to high-quality care can boost up the reduction of the mortality rate to 50%.²⁸

Women had higher rates of AMI in-hospital mortality. Studies from high-SDI countries found a higher risk for in-hospital mortality in women aged >60 years than in their male counterparts.^29,30 This bespeaks a lack of awareness about the risks and a poor recognition of the symptoms of women who have AMI. These are frequently older, suffer atypical symptoms, and present bleeding and vascular complications. Also, delays in medical care and a lower probability of receiving guideline-based pharmacological therapies and revascularization have been reported in women.^31,32,33

In 2018, the MOH launched the National Program to Reduce the Mortality due to AMI (PREMIA, Spanish acronym), which developed training materials for medical doctors to diagnose AMI and build clinical guidelines, encouraged the creation of health networks, and promoted pharmacological reperfusion centers, mainly in those states that lacked hemodynamic rooms.³⁴ However, the PREMIA program lacks impact evaluations or progress reports.

The competence of MOH 2^ndand 3^rd level hospitals to diagnose AMI and perform pharmacological and mechanical reperfusion is poor, and there are wide gaps among the states. These results signal critical disparities in the strategic planning of healthcare services for patients with AMI. It is possible to assume that the MOH hospital network performance for AMI treatment has been substandard, since the AMI in-hospital mortality increased during the period of the analysis.

Deficiencies on the supply side might explain the disparities in the AMI in-hospital mortality. We found that half of the hospitals had shortages for diagnosing AMI. The EKG recording and interpretation at first medical contact of the patient with AMI symptoms is a primary tool in the early identification and management of AMI,³⁵ that allows prompt diagnosis, especially in low resource settings. Additionally, few hospitals could perform pharmacological and mechanical reperfusion. Reperfusion therapy in ST-segment elevation AMI (STEMI) is an essential component of the treatment, as it strongly influences short- and long-term patient outcomes.³⁶ Primary mechanical reperfusion is recommended as the preferred therapeutic reperfusion strategy for all patients with STEMI when it can be performed within 90 mins. After first medical contact, or for patients with contraindications to fibrinolytic drugs, or those in cardiogenic shock; otherwise, pharmacological reperfusion with thrombolytics is crucial. However, we found that only 27% of AMI patients received thrombolysis, and a mere 12% underwent angioplasty. These figures are lower than in other countries. For instance, two prospective studies of acute coronary syndrome survivors from 20 European, Latin American, and eight Asian countries reported that the mechanical reperfusion rate ranged between 24.8% (in India) and 65.6% (in Northern Europe), and fibrinolysis rates, between 8.1% (in China) and 34.2% (in Southeast Asia), which resulted in total reperfusion therapy ranging between 53.9% (India) and 81.2% (Southern Europe).³⁷

The study has limitations: first, the analysis used routinely collected data on hospital discharges from the Health Information System of the MOH. The quality of these data regarding the accuracy of the diagnosis of AMI has not been validated in Mexico. Studies from high-SDI countries showed that hospital discharge data are reliable as to the diagnosis of AMI.^38,39 Second, the starting point of our analysis of in-hospital mortality was 2005, or the nearest year with a registry. Third, we estimated in-hospital mortality rates adjusting by age and sex, but we were unable to analyze comorbidity as the cause of death. Fourth, we could not ascertain the competence of the MOH to provide pre-hospital care; therefore, we were unable to identify the barriers to the emergency transportation of patients that were seen in emergency services at outpatient care clinics. Fifth, we were unable to identify patients who were discharged and then re-hospitalized on the same day, which can lead to overestimation of the hospitalization rate.