Disability-adjusted life years due to COVID-19 in Sri Lanka: a retrospective cross-sectional study

Introduction

COVID-19 has rapidly spread throughout the world in a very short period of time.1 By the end of 2022, 665 543 700 COVID-19 cases and 6 702 398 deaths due to COVID-19 were reported in multiple waves with the emergence of mutant variants, being second only to the 1918 Spanish influenza pandemic, despite the development of effective vaccines that were approved by WHO for emergency use.2 3 The morbidity and mortality due to COVID-19 infection are probably underestimates given that reporting and death registration systems in many countries are not 100% accurate and complete. In addition to causing a health burden, the COVID-19 pandemic has led to social and economic consequences throughout the world.

The mortality and development of complications from the diseases vary according to the age and presence of other comorbidities of affected individuals. Evidence shows that individuals who are older and having other comorbidities such as diabetes, cardiovascular diseases and chronic respiratory diseases are at increased risk of developing severe health consequences of COVID-19 infection.4 5

The health impact of COVID-19 has been investigated globally. The effects of the disease have been measured using mortality-based metrices such as case fatality rate, crude mortality rate and all-cause mortality. The number of deaths in a given region is never an accurate measure of mortality and does not reflect the true burden of the disease. To overcome this drawback, disability-adjusted life year (DALY) metric, developed by Murray and Lopez in 1994, is used instead. DALY is a composite measure which is calculated by taking the sum of two components: years of life lost (YLL) due to premature death and years lived with disability (YLDs).6 DALYs were originally used in the first Global Burden of Disease (GBD) study which quantified the health effects of >100 diseases for eight regions in the world in 1990.7 The advantage of using DALYs to measure disease burden is that it accounts for both mortality and morbidity associated with diseases which gives a more comprehensive assessment of the state of health in communities. However, the main disadvantage of DALYs is that they focus solely on health and do not capture the broader societal impact of diseases.8 Nevertheless, DALYs are commonly used to measure disease burden, and due to their wide usage, it enables comparison of disease burden between diseases, countries and different communities.

The first case of COVID-19 in Sri Lanka was detected on 27 January 2020 in a female tourist aged 44 years from Hubei province, China.9 The second case, which was also the first local case, was detected on 10 March 2020 in a tour guide aged 52 years who had recently been exposed to an Italian tour group.9 Sri Lanka was able to effectively curb the spread of the virus with rigorous measures for several months. However, during the first year of the pandemic, the epidemic spread originating from four clusters that were identified probably due to the gradual lessening of stringency measures with time to kick start a failing economy.

The Epidemiology Unit of the Ministry of Health, Sri Lanka maintains morbidity and mortality statistics related to COVID-19. The aim of this study was to estimate the DALYs related to COVID-19 pandemic in Sri Lanka from 27 January 2020 to 30 June 2022 including a component that accounts for the psychological stress that contacts may have experienced due to the stringent measures that were adopted during this time.

Materials and methods

Results

A total of 664 123 COVID-19 cases (males 54%) and 16 521 deaths due to COVID-19 (44% males) were reported in Sri Lanka from 27 January 2020 to 30 June 2022 (table 1). The case fatality rate increased with age, each 10-year group having a higher case fatality rate than the preceding younger age group. Females had a significantly higher case fatality rate in all age categories except in the extremes of age (figure 1). Almost one-quarter of infected persons above 80 years died (table 1). The total years of life lost during this period is estimated to be 103 568 for males and 159 880 for females.

Request permission

Males lived 1825 years with disability during this period and females 1531 years based on a disability weight of 0.133 (as for a severe lower respiratory tract infection assuming that the disability was for a period of 2 weeks (0.038 years)) (table 2).

Given the strict regulations imposed during the pandemic, all contacts were quarantined for 14 days; we assumed that all household members where a COVID-19 case was detected would have quarantined for 14 days where they were likely to have been anxious. Assuming that the average Sri Lankan household has 3.7 persons18 and using a disability weight of 0.03 (as for mild anxiety)17 for a period of 14 days (0.038 years), we estimated the YLDs for the contacts to be 2801.

Table 3 gives the summary of the DALYs due to COVID-19 in Sri Lanka from 27 January 2020 to 30 June 2022. An estimated total of 269 606 DALYs was experienced by the Sri Lankan population of 22 159 542 during this period. The DALYs per 1000 population increased with age and was higher in females than in males except in the extremes of age groups (table 3 and figure 2).

Request permission

Discussion

We have estimated that a total of 269 606 DALYs were lost due to COVID-19 in Sri Lanka between 27 January 2020 and 30 June 2022, the number being higher among females as compared with males primarily due to a higher case fatality rate. As expected, the number DALYs per 1000 population was higher among females except at the extremes of age.

In our study, almost 12 DALYs were lost per 1000 population. In Italy, it has been reported that since the inception of the pandemic in 2020 until the 28 April, 2020, 2.01 DALYs were lost per 1000 population.20 In Iran, the estimated loss of DALYs per 1000 population in 2020 was 22.2 (table 4).21 Our estimates are for over a 2.5-year period unlike the estimates for Italy and Iran, which are for shorter periods of time; it is likely that their estimates for a similar period of 2.5 years will be much higher than that for Sri Lanka which we report here.

One reason for the relatively low DALY counts we report here may be due to Sri Lanka being an island nation which limits travel and the stringent measures taken by the government to control the spread of the disease. Initially, there were curfew type lockdowns and major travel restrictions imposed. All contacts were actively tracked down and tested for COVID-19. Till about the end of June 2022, all contacts were required to quarantine or be on home management for 2 weeks. Despite these measures, Sri Lanka experienced three waves of infection during this period; the impact of each wave was more severe than the previous one and the case fatality rate was higher in each succeeding wave.

Sri Lanka started the vaccination campaign in late January 2021 with limited stocks. The programme was expanded in the next few weeks with the availability of more vaccines. Although the uptake of the first and second doses, even though the majority was vaccinated with the Chinese Sinopharm vaccine, was relatively good, the uptake for the third and fourth doses was poor despite the availability of the Pfizer-BioNTech COVID-19 Vaccine. The partially successful vaccination programme is also likely to have contributed to the low DALY estimates.

We have included an estimate of the YLD for contacts for the first time in estimating DALYs. The reason for doing so is the stringent control measures that were implemented during this period. As contacts were required to quarantine or stay home for 14 days, we postulated that they would have suffered some amount of anxiety during this period; some of the quarantine centres were poorly equipped with minimal comforts and were situated far away. For purposes of estimations, we assumed a disability weight of 0.03, the same as for mild anxiety, for a period of 2 weeks for all household contacts estimated, on average, to be 3.7 per household. Deaths were included in the same calculation and their YLDs were considered to be the same as for non-fatal infections.

Despite the inclusion of DALYs experienced by contacts, we acknowledge that our estimates are underestimates. We believe that the actual number of cases is more as some patients with typical signs and symptoms of infection did not seek medical care due to the stigma that was associated with being infected with COVID-19 even though the reporting system of diagnosed cases was very reliable. This is likely to have underestimated the DALY count for contacts as well.

We did not take into account DALYs lost due to long COVID-19 as this information was not available for Sri Lanka; if we did, our DALY estimate would have been much higher. In calculating DALYs, we did not account for age weighting and social discounting as recommended by WHO.14 However, we did not use 90 years as the average life span at birth in 2050 as it did not seem realistic for Sri Lanka; instead, we used life expectancies based on Sri Lankan demographic data.

YLDs do not measure the limitations of functioning, or social participation,22 which was the main type of disability experienced by the contacts due to mandatory social isolation and the stigma attached to the disease. As suggested by Grosse et al, measurement of limitations of functioning, physical activities and social participation derived from information gathered from those affected is essential to the development of meaningful measures of disability.22 In this study, we have attempted to calculate YLD of contacts by factoring in the anxiety they experienced. The impact of the disease may vary widely across countries and socio-economic strata as availability of resources and access to them vary widely across countries and among communities. Hence, even though the YLDs are comparatively smaller, their impact may have been more than that experienced in other settings.

The significantly higher case fatality rates among females in different age groups except at the extremes of ages were not expected. In Sri Lanka, there is no evidence of male preference or having access to better healthcare facilities. The literacy rate among both males and females is high at over 94%. Almost all deliveries take place in hospitals having specialised services and personnel. Females comprise about 60% of university admissions and >60% using healthy lifestyle clinics for persons 40 years and older are females. Grosse et al reported that the case fatality rate for COVID-19 was higher among males than among females,22 a finding which was consistent across many countries for several age groups. In Israel, a study reported that females were 1.12 times more likely to be infected; however, mortality was higher among males. The observation that more males being infected but the case fatality rate being higher in females needs further investigation.

Although the DALY counts due to the COVID-19 pandemic in Sri Lanka we report here are low and are probably an underestimate, resource availability and other impacts on daily life should be taken into consideration when interpreting the results and the overall impact of the pandemic.