Need for improving COVID-19 mortality registries: the case of Peru

Abstract

To the Editor: The SARS-CoV-2 pandemic has generated unprecedented health consequences with a record of nearly 117 million cases and more than two million deaths in 188 countries as of July 2020 [1]. Peru was the 19th country in the world and fifth in Latin America [1] with the highest number of identified cases (1.5 million). In addition, there are 50 thousand deaths [2], and an excess of 355 deaths per 100,000 inhabitants by 2020 attributed to this condition compared to previous years [3]. Correct recording and monitoring of the dynamics of COVID-19 mortality are important to know the impact of the pandemic and the effectiveness of health measures, as well as to allow timely rethinking of these measures if necessary. However, under-registration of mortality has been identified in most Latin American countries, including Peru [4]. In Peru, the main sources of information on COVID-19 mortality are the National Informatics Deaths System (SINADEF in Spanish) and the Situational Room of the Ministry of Health of Peru (MINSA in Spanish). SINADEF is a virtual information system, in which medical personnel record the causes (basic, intermediate, direct, or intervening) of the death of patients, including those with suspected or confirmed COVID-19 cases [5]. In turn, the Situational Room of MINSA takes the National Center for Epidemiology, Prevention and Disease Control (CDC) as a source of information, which centralizes reports of all cases of death by COVID-19 verified by a laboratory test, recorded by health services at the national level through its systems of epidemiological surveillance and health intelligence [6]. According to the World Health Organization (WHO), for epidemiological surveillance purposes, death by COVID-19 is considered to occur as a result of a clinical picture compatible with COVID-19, whether a suspected or confirmed case [7]. Likewise, the Pan American Health Organization (PAHO) points out that the codes U07.1 (identified virus) and U07.2 (unidentified virus) of the International Classification of Diseases 10th edition (ICD-10) should be used to record the cases of deaths due to this disease [8]. Despite these recommendations, in July 2020, the Situation Room of MINSA reported the presence of an under-registration of more than 3000 deaths in the period from March to July 2020 compared to the SINADEF records [9]. This reflects a major problem that does not allow assessment of the actual impact of the pandemic according to the recommendations of the WHO and PAHO, generating confusion among the population and decision-makers. To determine the extent of this under-registration throughout the national territory and to know where this problem is concentrated, an ecological study was carried out to estimate mortality by COVID-19 at the departmental level from codes U07.1 and U07.2 recorded in SINA DEF (https://bit.ly/3hjRbOA) from the first reported case of COVID-19 in Peru to March 6, 2021, and compare the results with the figures reported by the Situational Room of MINSA in the same period (https://covid19. minsa.gob.pe/sala_situacional.asp). Furthermore, the code enabling the replicability of results in the R statistical program is available on the GitHub collaborative developmen t p l a t f o rm ( h t t p s : / / g i t h u b . c om / j e g un s / ComparacionMortalidad). A total of 94,574 COVID-19 deaths recorded under the ICD-10 codes were identified: U07.1 (n = 70,099) and U07.2 (n = 24,475) in SINADEF, while the Situational Room recorded 47,491 deaths by the same pathology (a difference of 47,088 deaths). This inconsistency