Introduction

The COVID-19 disease pandemic is a global health emergency caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The first human case of this disease occurred in late 2019 and in four months it spread to almost every country in the world ^1,2.

The majority of the COVID-19 infected population experiences mild to moderate respiratory disease and recovers without requiring special treatment. Older people and those with underlying medical problems such as chronic non-communicable diseases (cardiovascular, diabetes, chronic respiratory diseases, and cancer) are more likely to develop serious diseases, require ventilatory support, and die from complications ^3-7.

The actual number of deaths from SARS-CoV-2/COVID19 is underestimated because hospitals, medical providers, and health authorities are reporting only confirmed cases. The effective number of deaths caused by the SARS-CoV-2 virus will only be evident after statistically modelling excess mortality during the period that this pandemic last.

Through collaborative inter-institutional work between the Municipal Public Health Secretary, the Hospital Registries of Cali, the epidemiological surveillance system for childhood cancer VIGICANCER and the Population Cancer Registry of Cali, the behavior of cancer patients will be monitored to see the impact of the COVID-19 pandemic. The objective is to establish the baseline to calculate excess mortality during the SARS-Cov-2 pandemic. The method can be used in other regions or cities that have registries and observatories of chronic non-communicable diseases and cancer.

Materials and Methods

Cali is the capital of Valle del Cauca Department and the third largest city in Colombia. According to the 2018 census and according to DANE projections, the estimated population for 2023 is 2.3 million inhabitants of which 54% will be women ^8,9. Of the total population of Cali, 26.2% recognize themselves as belonging to the black ethnic group ¹⁰. Life expectancy at birth is 74.4 years old ¹¹ with gender distribution of 82.7 for women and 77.4 for men ¹². The infrastructure for cancer care has 165 authorized cancer services ¹³. These services are found in the urban area where 95% of the population lives. That population lives in an area of 110 km², which corresponds to 20% of the extension of the municipality of Cali (503 km²).

Information on the number of deaths, by basic cause from January 2003 to February 2020, was obtained from the general mortality database of the Municipal Health Secretariat of Santiago de Cali. The methods for its epidemiological registration have been previously described ^10,14. The International Classification of Diseases (ICD-10) ¹⁵ was used to code deaths, and following the guidelines of the World Health Organization (WHO) ¹⁶⁾ the causes of death were grouped into three major groups of diseases: communicable, chronic, non-communicable and injuries; the rubrics of each category are detailed in Table 1.

Deaths from respiratory infections and chronic non-communicable diseases associated with fatal outcome during the COVID-19 pandemic (malignancies, diabetes mellitus, and cardiovascular and respiratory diseases) were included in the analysis. The structure of the population by five-year age groups for each calendar year was obtained from the National Administrative Department of Statistics of Colombia (DANE) ¹⁷. Mortality rates for the entire population were age-standardized (ASR) by the direct method using the world standard population as a reference. Global and age-specific rates are expressed per 100,000 people-years. The trend in mortality rates between 2003 and 2009 was described using the Annual Percent Change (APC), calculated using the weighted least squares method ¹⁸. To detect seasonal changes, monthly mortality rates were estimated during the 206 months evaluated (January 2003 to February 2020).

Results

During the period 2003-2019 there were 207,261 deaths and 65,906 in the five-year period 2015-2019, with the following distribution of frequencies: Communicable diseases (7,249; 11.0%), chronic non-communicable diseases (50,121; 76.0%), and injuries (8,536; 13.0 %). For the diseases included in the analysis, the frequency distribution was as follows: Respiratory infections (6.1%), respiratory diseases (5.4%), cancer (22.3%), diabetes (2.6%) and cardiovascular disease (30.1%).

The average annual mortality rate per 100,000 people-year for all causes of death was higher among men (568) than among women (322). Excess mortality could be explained by chronic non-communicable diseases such as malignant neoplasms and cardiovascular diseases, and by intentional injuries (Table 2).

The mortality rates from chronic non-communicable diseases in the five-year period (2015-2019) were like those in 2019 (Table 4-Suppl); however, mortality rates from communicable diseases and injuries were lower in 2019 compared to those observed in the five-year period (2015-2019) (Table 2).

Figure 1 shows that the risk of dying is directly associated with aging and that mortality rates from cardiovascular diseases remained stable in people older than 65 years old. In contrast, a significant decrease in the risk of dying was observed between 2003 and 2019 among those younger than 65 and in all age groups, related to respiratory diseases and diabetes mellitus. The magnitude of the decrease was greater in the group of 50-64 years old. Influenza and pneumonia death rates decreased significantly in children younger than 5 years old and remained stable in the other age groups.

APC: Annual Percent; CI: 95% confidence interval during the study period (17 years). * APC is significantly different from zero (0)

Figure 1 Cali, Colombia. Trend in age-specific mortality rates for some selected categories during the period 2003-2019. Mortality rates are expressed per 100,000 people per year. 

Figure 2 shows the temporal variation of the monthly age-specific mortality rates for the respiratory diseases group and the monthly variations of the global rates for the cardiovascular disease, cancer and diabetes mellitus groups. Age-specific rates were highest in those older than 80 years for all diseases evaluated. In the group of respiratory diseases, influenza and pneumonia, there was more evident seasonal variation in the elderly and with few exceptions, the peaks occurred between the months of October and January.

Figure 2 Cali, Colombia. Changes in monthly general and age-specific mortality rates for respiratory infections and selected chronic non-communicable diseases between January 2003 and February 2020. In deaths from respiratory diseases (influenza and pneumonia), the monthly mortality rates per 100,000 people-year show seasonal variation and with few exceptions, the peaks occurred between October and January. 

Discussion

Estimates of the magnitude and temporal trend of mortality rates from respiratory infections and chronic non-communicable diseases for the entire population of Cali over three five-year periods, provide a robust baseline that will serve as a comparison to estimate the excess mortality caused by COVID-19 pandemic. Seasonal variations in mortality were evident in the group of older adults suggesting that this section of the population bears the heaviest burden in terms of the severity of seasonal influenza. Patients suffering from COVID-19 that have the highest risk of death are those of advanced age and with the presence of pre-existing conditions such as diabetes mellitus, cardiovascular diseases, cancer, hypertension and smoking. The impact of the disease on them can be measured by the case fatality rate (CFR) ¹⁹.

Properly classifying the causes of death during the COVID-19 pandemic is an emerging challenge. It is imperative that health authorities and decision makers be guided by reliable estimates of mortality and IFC. This index measures the proportion of infected people who die from SARS-CoV-2 virus infection. An accurate count of the number of deaths due to COVID-19 depends in part on proper completion of the death certificate ²⁰. When a patient suffering from COVID-19 dies, it is likely that this disease is the basic cause of death, even in patients with comorbidities such as chronic, non-communicable diseases, which are contributing causes, but are not part of the causal sequence.

The COVID-19 CFR shows the pattern of an emerging infectious disease. At the beginning of the disease and when first cases are described, CFR is high and then decreases as the pandemic progresses. The denominator of the CFR in the COVID-19 pandemic is variable and closely related to the control policies implemented by regional and/or national health authorities ¹⁹. A major challenge in accurately calculating IFC is the denominator: the number of people infected with the virus. Asymptomatic cases of COVID-19, patients with mild symptoms, or individuals misdiagnosed (false negatives) could be left out of the denominator, leading to underestimation of the number of infected and an overestimation of CFR ¹⁹.

CFR depends on the number of tests performed to detect people with COVID-19 disease. The countries that achieved control of the pandemic have a higher denominator and a lower CFR because at the beginning of the pandemic they invested resources in mass analysis with intensive tracing of cases and tests on contacts, without limiting themselves to seriously ill patients. Their proposals included the implementation of extreme social isolation, localized quarantines, and follow-up monitoring of suspected cases, even in vulnerable populations, to contain the spread. In contrast, countries that did not exercise control measures for the pandemic have a lower denominator and a high CFR because in the presence of dozens of infections and multiple possible asymptomatic cases, confirmatory laboratory tests for the virus were reserved for patients with severe symptoms ²⁰. The lack of opportunity in the implementation of containment measures led these countries and cities to a collapse in health services due to the avalanche of critical cases, the inadequate operation of local laboratories, damming of samples, and a shortage of mechanical ventilators and health personal. In addition, they have also demonstrated a lack of diagnostic capacity that has made it impossible to truly know the dimension and impact of the disease.

The national percentage of duly certified deaths in Colombia is 93.7% ⁽²¹. To guarantee the validity of the estimates of excess mortality caused by COVID-19, it is a priority to standardize the basic cause of death during the pandemic. The WHO used the international disease classification (IDC) and assigned emergency headings U07.1 and U07.2 to deaths from SARS-CoV-2 infection confirmed by laboratory and clinically and epidemiologically diagnosed, respectively. Codes U00-U49 are used by WHO for the provisional allocation of new diseases of uncertain aetiology. In emergency situations, codes are not always accessible in electronic systems. Specification of category U07 in the manner in which it is done in Table 3 will ensure that this category and its subcategories are available in all electronic systems at any time and that they can be used immediately according to WHO instructions ²².

The collapse of health systems and even funeral services and the fear of becoming ill can contribute to all deaths occurring outside the hospital setting, as well as that of any symptomatic respiratory patient without confirmatory test, being directly attributed to SARSCov-2/COVID-19. This current problem facilitates the production of cases of poor classification when registering the causes of death in death certificates. This will cause a trend that may suggest a false decrease in mortality rates attributable to chronic non-communicable diseases, specifically those related to cancer.

Comments

The massification of diagnostic tests, the reduction in response time and the mapping of serious respiratory infections are the main strategies used by developed countries to slow down the spread of the disease. In countries where hundreds of deaths are recorded daily, where there are no diagnostic tests, and in those where the fear of being infected is greater, it is impossible for health sectors to accurately diagnose the cause of death.

Knowing the mortality rates, their time trend and the frequency distribution of deaths in Cali before the pandemic, will allow modelling excess mortality to determine the real impact of the disease.