Descriptive epidemiology of SARS-CoV-2 infection in Karnataka state, South India: Transmission dynamics of symptomatic vs. asymptomatic infections.

BACKGROUND: The huge surge in COVID-19 cases in Karnataka state, India, during early phase of the pandemic especially following return of residents from other states and countries required investigation with respect to transmission dynamics, clinical status, demographics, comorbidities and mortality. Knowledge on the role of symptomatic and asymptomatic cases in transmission of SARS-CoV-2 was not available.
METHODS: The study included all the cases reported from March 8 - May 31, 2020. Individuals with a history of international or domestic travel from high burden states, Influenza-like Illness or Severe Acute Respiratory Illness and high-risk contacts of COVID-19 cases were included. Detailed analysis based on contact tracing data available from the line-list of state surveillance unit was performed using cluster network analysis software.
FINDINGS: Amongst the 3404 COVID-19 positive cases, 3096 (91%) were asymptomatic while 308 (9%) were symptomatic. Majority of asymptomatic cases were in the age range of 16 and 45 years while symptomatic cases were between 31 and 65 years. Mortality rate was especially higher among middle-aged and elderly cases with co-morbidities, 34/38 (89·4%). Cluster network analysis of 822 cases indicated that the secondary attack rate, size of the cluster and superspreading events were higher when the source case was symptomatic as compared to an asymptomatic.
INTERPRETATION: Our findings indicate that both asymptomatic and symptomatic SARS-CoV-2 cases transmit the infection, although symptomatic cases were the main driving force within the state during the beginning of the pandemic. Considering the large proportion of asymptomatic cases, their ability to spread infection cannot be overlooked. Notwithstanding the limitations and bias in identifying asymptomatic cases, the findings have major implications for testing policies. Active search, early testing and treatment of symptomatic elderly patients with comorbidities should be prioritized for containing the spread of COVID-19 and reducing mortality. FUNDING: Intermediate Fellowship, Wellcome Trust-DBT India Alliance to Giridhara R Babu, Grant number: IA/CPHI/14/1/501499.

Evidence before this study

Understanding transmission dynamics in the spread of SARS-CoV-2 is important for controlling the COVID-19 pandemic. We retrieved all English-language articles reporting the transmission dynamics of SARS-CoV-2 from symptomatic and asymptomatic cases through systematic searches of major databases that included PubMed, medRxiv, bioRxiv, arXiv, SSRN, Research Square and Wellcome Open Research for peer reviewed articles, preprints and research reports. Our search identified publications that included terms related to ‘symptomatic’, ‘asymptomatic’, ‘transmission dynamics’, ‘covid-19′and ‘SARS-CoV-2′ between January 1 - June 1, 2020. Data pertaining to the role of symptomatic and asymptomatic cases in transmission of SARS-CoV-2 during early phase of the pandemic was not available from India or any LMIC with a similar setting.

Added value of this study

The study provides evidence that: (i) Transmission of SARS-CoV-2 occurs both from asymptomatic and symptomatic cases. (ii) Symptomatic index case results in higher secondary attack rate, bigger size of the cluster and higher occurrence of overt clinical illness. (iii) SSE's are more commonly associated with symptomatic index cases. (iv) COVID-19 mortality is higher in elderly with comorbidities.

Implications of all the available evidence

The findings of this study have major implications for policies related to surveillance and testing. Active search and early detection of symptomatic elderly patients with comorbidities should be prioritized to reduce mortality. Further, testing of all individuals in high risk settings to capture silent transmission by asymptomatic individuals should be an integral component of the policy to prevent spread of SARS-CoV-2.

Alt-text: Unlabelled box

Introduction

India reported the first case of coronavirus disease 2019 (COVID-19) on January 30, 2020, in a medical student who travelled from Wuhan, China [1]. In the beginning, COVID-19 cases in India were associated with international travel among returnees from COVID-19 affected countries. Since then, Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection has spread across all states. As of September 16, 2020, a total of 5020,359 SARS-CoV-2 infections have been reported from India [2]. Karnataka, a state in the southern part of India, has a population of 64·41 million spread across 30 administrative units called districts. Karnataka state reported the first case of COVID-19 in an international traveler on March 8, 2020 [3]. In the subsequent three months (March 8 to May 31, 2020) the number of COVID-19 cases steadily increased to 3404 in the state. The health authorities of the state initiated laboratory-based surveillance, traditional infection control and public health strategies to contain the spread [4,5]. These included tracing, testing and tracking of contacts and isolating COVID-19 positive subjects [6]. The state implemented several public health measures, including retrospective contact tracing, survey for influenza-like illness and strengthening the laboratory network under the Integrated Disease Surveillance Programme (IDSP). Also, the state had more than 17 task forces set up under the Karnataka State Disaster Management [7]. Several innovative measures were taken to tackle COVID-19 in the initial months. For example, invoking powers under Epidemic Diseases Control Act, the state started using information from available apps to notify people about their possible interaction with a COVID-19 positive person.

From March 25, 2020, the country was under complete lockdown until May 31, 2020, in four phases [8]. The first phase of the lockdown was from March 25, 2020, to April 14, 2020. The second phase of lockdown was from April 15, 2020, to May 03, 2020, the third phase of lockdown from May 04–17, and the fourth phase from May 18–31, 2020. From the beginning of the third phase, the Government of India issued guidelines permitting inter-district and inter-state travel for stranded persons. Consequently, there was a huge surge of COVID-19 positive cases from May 4–31, by which time most of the residents of Karnataka (n = 93,073) from other states, especially from those with a high burden of COVID-19 cases, returned home [9]. As of September 16, 2020, a total of 8,932,699 samples have been tested in the state, of which 851,212 tested positive (9·5%) [10].

From the beginning of the pandemic, there has been considerable debate on the role of asymptomatic and symptomatic persons in spreading the SARS-CoV-2 infection. Initially, the World Health Organization (WHO) inferred that the role of asymptomatic persons in transmission was minimal [11,12]. However, the WHO acknowledged later that there is a growing body of evidence that even asymptomatic persons can spread the disease [13]. Previous reports suggest that nearly 4–32% of the infections are spread through asymptomatic persons [14,15]. Determining the role of asymptomatic persons in viral transmission in low and middle-income countries (LMICs) is very important. The urban areas in these countries have high population density, poor personal hygiene, with ubiquitous crowding. In this report, we describe the epidemiology of SARS-CoV-2 infection in the southern state of Karnataka, India, during the lockdown phase with special emphasis on transmission dynamics of symptomatic and asymptomatic cases.

Materials and methods

Study population

The study included all cases diagnosed with COVID-19 in the Karnataka state, reported from March 8 - May 31, 2020. Case identification was carried out by the staff of each district, which included the district surveillance officer and their teams, the Integrated Disease Surveillance Programme (IDSP) team, and the urban health authorities of the major cities. Reporting of Influenza-Like Illness (ILI) and Severe Acute Respiratory Illness (SARI) cases was mandatory for even private health care providers. The contacts of COVID-19 positive people who were in quarantine were monitored. With respect to domestic and international travellers, the state had a policy of quarantining and subsequent testing of all passengers exposed to a COVID-19 positive case. Most of the travellers were from COVID-19 affected countries or states within India. Upon arrival in to the state, they were immediately subjected to testing. If tested positive they were hospitalized and if negative, were quarantined for 14 days with a repeat test being carried out between 7th and 10th day of quarantine [16].

Case definitions

A suspect case of COVID-19 was essentially based on the criteria defined by the Ministry of Health and Family Welfare, Government of India described elsewhere [17]. Briefly, any person with a recent history of international travel (14 days), domestic travel from high burden states of Maharashtra, Gujarat, Delhi, and Tamil Nadu and anyone within the state with symptoms of ILI and SARI as well as known high-risk contacts of a confirmed COVID-19 patients were also included. Asymptomatic SARS‐CoV‐2 cases were defined as those with positive SARS‐CoV‐2 RT-PCR in the absence of symptoms [18].

Laboratory testing

Testing for SARS-CoV-2 was done initially at five laboratories in the state. This was quickly ramped up to 63 laboratories across the state by May 31, 2020, resulting in an enhanced capacity for testing and identification of COVID-19 cases across Karnataka. Nasopharyngeal and oropharyngeal swabs collected into virus transport medium from suspected cases were subjected to an RT-PCR for detection of SARS-CoV-2 RNA using the Indian Council of Medical Research (ICMR) guidelines for testing [19]. Three RT-PCR kits approved by ICMR were used in the study and the details are provided in Table S1. All positive and negative results were entered immediately upon their availability into the ICMR portal and shared with the district/state surveillance teams to facilitate immediate tracing of contacts.

Data collection

All the sociodemographic, clinical, and risk factor details (travel history, symptoms, comorbidities, etc.) of patients for every suspected case was collected on the day of sampling and entered into a standardized line list by the surveillance teams in the respective districts/cities. The line list also included information about the type of sample, date of sample collection, date of testing, type of RT PCR kit used, and the test results, as well as the details of contacts. All contacts were traced based on the information provided by the individual and subjected to testing as per the national guidelines [20]. All the data was maintained in a centralized database by the State Surveillance Officer. Every effort was made to ensure that missing data, if any, was collected from the subjects by the surveillance team. All the data were de-identified before extraction and analysis. A subject who was detected SARS-CoV-2 positive was given a P number. The lesser the P number, the earlier the patient was detected either as a source or as a contact of a known source. The study has been reported in accordance with the RECORD guidelines.

Data analysis

The data analysis from the line list was based on the date of collection of a sample from March 8 to May 31, 2020. The trends in the positivity rate over time were described by calculating the seven-day moving average. The number tested per million was computed by using a State-specific population denominator. The number of individual tests and contacts tested per confirmed case was calculated. Laboratory accessibility was examined by subtracting the date of specimen collection from the specimen receipt date. An indicator for contacts tested per case was estimated by dividing the number of tests among contacts of cases with the number of positive cases. Contact tracing information available in the line list was analyzed to determine the source of infection. Based on the contact tracing information, cluster network analysis was performed and presented using Microsoft Excel and Gephi network analysis software [21]. Secondary attack rates were calculated using standard methods, and the dispersion parameter was calculated for asymptomatic and symptomatic subjects and expressed as the k value [22,23]. As suggested recently [24] smaller values of k indicate greater heterogeneity, and in general, k value<1 is associated with a high number of ‘superspreaders’.

Descriptive epidemiology of cases by time, place and person

The frequencies of characteristics of cases were described by age, gender, residence, type of exposure (contact or travel), and symptoms. The presence of any symptoms at the time of specimen collection was also recorded. The date of specimen collection was used to draw the epidemic curve. The time trends were annotated with that of the implementation of various public health measures or key events related to the epidemic (Fig. 1).

Fig. 1

Distribution of case categories, number of cases and interventions undertaken. X axis depicts the timeline while Y axis depicts the number of COVID-19 positive cases. The date wise interventions undertaken by the state of Karnataka is described. The category of patients and the progression in numbers is depicted in various colors in a wave pattern.

Statistical analysis

All the data were entered into the excel sheet and analyzed using the Statistical Package for Social Sciences (SPSS, version 26) to determine the significance of various parameters. The cluster network analysis was performed using Gephi (version 0·9·2) for analyzing the transmission dynamics. Statistical significance was expressed as a p-value, calculated using the chi-square test, t-test for 95% confidence interval (CI).

Funding

Giridhara R Babu is funded by an Intermediate Fellowship by the Wellcome Trust DBT India Alliance (Clinical and Public Health research fellowship); grant number: IA/CPHI/14/1/501499.

Role of the funding agency

The funding agency of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the article. All the authors had full access to all the data in the study and corresponding author had final responsibility for submitting for publication.

Results

Testing

Figure S1 presents the district-wise distribution of cases. The highest number of COVID-19 cases were identified in the capital city of Bengaluru, followed by Kalburgi, Mandya, Yadgir, Udupi, and Raichur. During the study period, the Karnataka state was testing approximately 4377 per million, with a positivity rate of 1·1%. The total cases per million were 2708 in the state. Karnataka followed contact tracing diligently with the highest number of contacts (47·4) traced for each positive case detected [4], [18].

Description of cases by time, place, and person: The age, gender distribution, and clinical details of all the COVID-19 cases in Karnataka are presented in Table 1. To understand the dynamics of SARS-CoV-2 transmission within the state, we undertook a detailed analysis of the data available in the line list. Amongst the 3404 COVID-19 positive cases, 3096 (91%) were asymptomatic, while 308 (9%) were symptomatic. The age distribution revealed that 54·6% (1691/3096, 95% CI; 52·8–56·3) of the asymptomatic cases were less than 30 years of age compared to 25·3% (78/308, 95% CI; 20·4–30·1); of symptomatic in this age group. On the other hand, 43·5% (134/308, 95% CI; 38–49) of the symptomatic were over 46 years of age compared to 15·8% (489/3404, 95% CI; 14·5–17) of the asymptomatics in this age group. The percentage of symptomatic and asymptomatic in the age range 31–45 were more or less similar (31·2% and 29·6% respectively) as shown in Table 1. The gender distribution revealed a male preponderance in both asymptomatic and symptomatic cases.

Table 1

Age group and gender of COVID-19 cases (Mean age 31·26) (n = 3404).

Age group	Asymptomatic (%; 95% CI)	Symptomatica (%; 95% CI)
<15	542 (17·5; 16·2 - 18·8)	17 (5·5; 3·0 - 8·0)
16–30	1149 (37·1; 35·4 - 38·8)	61 (19·8; 15·3 - 24·3)
31–45	916 (29·6; 28·0 - 31·2)	96 (31·2; 26·0 - 36·4)
46–60	374 (12·1; 11·0 - 13·2)	72 (23·4; 18·7 - 28·1)
>60	115 (3·7; 3·0 - 4·4))	62 (20·1; 15·6 - 24·6)
Total	3096	308
Gender ratio (M:F)	1884 (69; 67·3–70·6):1212 (31; 29·3–32·6)	211 (61; 55·5–66·4): 97 (39; 33·5–44·4)

Clinical manifestations of symptomatic cases are provided in supplementary Table S2.

Table 2 presents the clinical status and outcome of the COVID-19 cases in Karnataka. Overall, amongst the 3404 cases, the final outcome was available for 1394 (41%) at the end of the study period (May 31, 2020) while the remaining 2010 (59%) were active cases. Amongst those, where the outcome was available 7·6% (106/1394, 95% CI; 6·2–9) had a variety of comorbidities (Table S3). Further, the occurrence of comorbidities was ten-fold higher in symptomatic cases, 29·1% (73/251, 95% CI; 23·3–34·6), as compared to asymptomatic cases, 2·9% (33/1143, 95% CI; 1·8–3·7). The instant case fatality rate recorded in Karnataka was 3·8% (54/1394, 95% CI; 2·8–4·8). Amongst those who died, 87% (47/54, 95% CI; 78–96) were symptomatic cases, while 13% (7/54, 95% CI; 4–22) were asymptomatic cases. Among the 38 symptomatic cases with comorbidities who died, 89·4% (34/38, 95% CI; 79·6–99·2) were among the middle-aged and elderly. The cause of death among the seven asymptomatic cases is presented in Table S4. Overall, 39% (1340/3404) of cases recovered and were discharged from the hospital, while 59% (2010/3404) were active cases as of May 31, 2020. (not presented in the table).

Table 2

Comorbidities in COVID19 cases.

Outcome	Asymptomatic (n = 1143)		Symptomatic (n = 251)		Total (%)
	With comorbidity (%)	Without comorbidity (%; 95% CI)	With comorbidity (%; 95% CI)	Without comorbidity (%; 95% CI)
Recovered	33 (100; NA)	1103 (99·4; 98·4 - 99·5)	35 (48; 36·5 - 59·4)	169 (95; 91·8 - 98·2)	1340 (96·1)
Died	0	7a (0·6; 0·2 - 1·0)	38b (52; 40·5 - 63·4)	9 (5; 1·8 - 8·2)	54 (3·9)
Total	33	1110	73	178	1394

NA-Not Applicable.

The causes of death in these 7 patients were unrelated to COVID-19; Myocardial infarction (3), Traumatic Brain Injury (2), Suicide (1), and Septic shock (1).

Among these 38 cases, 34 (89.4%) were above the age of 45 years.

Fig. 2 shows the distribution of overall percent test positivity rate and the source in 3137/3404 cases. The acquisition of SARS-CoV-2 infection could mostly be attributed to travel (2264, 72·1%); domestic travel (2136, 68%) and international (128, 4%) or as a known contact with a COVID-19 positive case (873, 27·8%). In the remaining 267 cases (7·8%), the precise reason behind the acquisition of infection could not be ascertained. Amongst these 267 cases, 111 (41·6%) presented with ILI / SARI with no information of contact with a COVID-19 positive case. In the remaining 156 (58·4%) cases, the source of infection was not traceable. However, the initial adjusted test positivity rate was 1% on March 8, 2020 which gradually declined to less than 1% until mid-April 2020, following which there was a steady increased to 10% until the end of the study (Figure S2).

Fig. 2

Distribution of case categories and overall test positivity. X axis depicts the timeline, Y axis depicts the overall test positivity rate while the Z axis depicts the number of COVID-19 positive cases. Note there was a huge surge in the number of COVID-19 positive cases in the second half of May 2020.

As evident from Fig. 3, COVID-19 positive cases were initially detected in the city of Bengaluru and subsequently spread to people in the other districts, either by contact in Bengaluru or by travelling to other districts. The other hotspot districts were Gulbarga, Mysore and Dakshina Kannada, which had an independent source of infection and local spread. The majority of the COVID-19 cases (68·5%; 2264/3304) had a history of either international (5·7%; 128/2264) or domestic travel (94·3% 2136/2264), which implies that this was the main mode of introduction of SARS-CoV-2 into the state. Amongst those with a history of international travel 41/128 arrived at the city of Bengaluru. Similarly 44/128 arrived into the other international airport situated at Mangalore city of Dakshin Kannada district and travelled to their hometowns. Domestic travel from other states of India to the districts of Yadgir, Mandya, Udupi, Raichur and Kalaburgi accounted for more than 200 COVID-19 positive cases. Additionally inter-district travel within the state from the city of Bengaluru to Davanagere district contributed to the spread of 124 COVID-19 positive cases. Overall, our analysis revealed that the city of Bengaluru had the highest number of COVID-19 positive cases (378), followed by Kalaburgi (307), Mandya (284), and Yadgir(284), respectively.

Fig. 3

Distribution and direction of transmission of COVID-19 cases in Karnataka State. Each dot represents a person with COVID-19. Primary cases are in red and secondary cases in blue. Blue colored dots connected to primary cases using lines. The dots not connected by lines represent independent introduction from elsewhere. Bengaluru city was the epicenter, had the highest density of COVID-19 cases, from where spread occurred to several districts in the state.

Amongst the 873 cases with a history of contact with a COVID-19 positive case, 822 could be epidemiologically linked to 144 source cases. The remaining 51/873 had a history of contact but could not be precisely linked to a known source. Using the information available in the line-list, cluster network analysis was performed to derive dynamics of transmission (Fig. 4 and Table 3) from symptomatic and asymptomatic index cases. Amongst the 144 index cases, 32% (46) contributed to large clusters (>5 cases). As evident from Table 3, 69% (32/46) index cases were symptomatic while 31% (14/46) were asymptomatic. The size of each cluster ranged from 5 to 58 cases per cluster for symptomatic index cases (Mean 17·03; 95% CI 15·7–18·4). On the other hand, the clusters resulting from asymptomatic cases range from 5 to 23 per cluster (Mean 8·1; 95% CI 7·2–9·1). Amongst the 545 cases originating from 32 symptomatic index cases, 41 were symptomatic, and 504 were asymptomatic. On the contrary, only 2/114 cases originating from 14 asymptomatic index cases manifested symptoms (Table 3). The dispersion parameter (k value) of asymptomatic cases was thrice that of symptomatic cases. Detailed cluster wise distribution of cases are mentioned in Table S6. Graphic representation of transmission dynamics in the large clusters (>5 cases) is presented in Fig. 4. Overall, among the 545 cases resulting from symptomatic source cases, 67.5% (368) cases were contributed by 12 superspreading events (SSEs) originating from symptomatic cases. Whereas, two SSEs originating from asymptomatic cases contributed to 16·7% (19/114) cases (Fig. 4). Among the total 705 (46 source and 659 secondary) cases which fall into clusters of 5 or more cases, 291 accounted for within household transmission. The mean number of cases per household was 3.06 (95%CI; 2.8–3.2) ranging between 2 and 8 cases per household. Further, there was no significant difference observed with respect to the number of transmission cycles between clusters resulting from symptomatic (Mean 1·47; 95% CI 1·15–1·79) and asymptomatic (Mean 1·18; 95% CI 0·93–1·42) index cases (data not presented).

Fig. 4

Cluster diagram depicting symptomatic and asymptomatic COVID-19 cases in the Karnataka state, March–May 2020. Each case is represented as a dot: red dot indicates a symptomatic case while yellow dots represent an asymptomatic case. The index case in each cluster is depicted as a larger dot and also indicated with the P number. The lesser the P number, the earlier the patient was detected either as a source or as a contact of a known source. Note there were 46 clusters with five or more cases. The index case of majority (32/46) of clusters was a symptomatic patient (red). Large open circles (depicted using interrupted lines) represent superspreading events (>6 cases originating from a source). Overall, among the 545 cases resulting from symptomatic source cases, 67.5% (368) cases were contributed by 12 SSEs originating from symptomatic cases. Whereas, two SSEs originating from asymptomatic cases contributed to 16.7% (19/114) cases.

Table 3

Cluster network analysisa depicting the clinical status of index cases and dispersion parameter of COVID19 cases.

Index case status	Distribution of index cases (n = 46)	Total number of cases associated with the index case (n = 659)	Clinical status of cases associated with the index case (n = 659)	Average number of cases per cluster (95% CI)	Dispersion parameter (k value)
Symptomatic	32 (69%)	545 (82·7%)	Symptomatic (n = 41)	17·03 (15·7–18·4)	1·2
			Asymptomatic(n = 504)
Asymptomatic	14 (31%)	114 (17·3%)	Symptomatic (n = 2)	8·1 (7·2–9·1)	3·6
			Asymptomatic (n = 112)

Restricted to large clusters (>5 cases).

Discussion

This study describes the epidemiology of SARS-CoV-2 infections in the state of Karnataka, South India. Before March 8, 2020, there were no COVID-19 cases in Karnataka. SARS-CoV-2 infection was introduced into the state by people travelling from foreign countries until the country went into lockdown in the fourth week of March 2020 when international flights were not allowed into the country (Fig. 1). What initially started as an infection in the capital city of Bengaluru at the beginning of March 2020, quickly spread to the other parts of the state (Fig. 3). The majority of cases were mainly (68·5%; 2264/3304) attributable to domestic (94·3%, 2136/2264) and international travel (5·7%; 128/2264), thereby suggesting that this was the main mode of introduction of SARS-CoV-2 into the state. It is presumed that these subjects had acquired the infection at the place where they initiated their travel. Alternatively, it is possible that they acquired it during their travel from fellow passengers, especially the domestic travellers.

From the detailed epidemiological analysis of data in this study, it emerges that both symptomatic and asymptomatic cases contributed to the transmission and spread of SARS-CoV-2 infection within the state. An overwhelming majority of the COVID-19 cases in Karnataka were asymptomatic at the time of specimen collection, and this is similar to the results reported in India as well as other Asian countries [25] and in stark contrast to what is reported in several western countries [26]. Most of the asymptomatic cases in this study were young adults or middle-aged (Table 1: <46 years of age).

Our analysis suggests that the symptomatic cases were the prime drivers of the SARS-CoV-2 transmission within the state during the early phase of the pandemic. The evidence for this is manifold: (i) The mean secondary attack rate was 7·1 for symptomatic cases (90 symptomatic source cases transmitted infection to 645) as compared to 3·1 for asymptomatic cases (54 asymptomatic source cases transmitted infection to 177), and this difference was statistically significant (Table S5). (ii) Cluster network analysis revealed that symptomatic source patients contributed to 32 large clusters (>5 cases) with the mean size of each cluster being 17·03 (range 5–58). In contrast, the asymptomatic source cases contributed to 14 large clusters (Table 3 and Fig. 4) with the mean size of each cluster being 8·1 (range 5–23). (iii) The proportion of secondary cases in the clusters manifesting overt clinical illness was higher when the index case was symptomatic compared to asymptomatic (Table 3 and Table S6). (iv) The dispersion parameter for symptomatic source cases was three-fold lower (k = 1·2) as compared to asymptomatic source cases (k = 3·6), thereby, indicating the higher propensity of transmission from symptomatic cases especially associated with SSEs (>6 cases originating from an index case) [27].

Studies from western countries have observed that the true proportion of asymptomatic cases is as low as 20% [28]. On the contrary, several other studies have observed higher proportion of asymptomatic cases especially in India [19,[29], [30], [31], [32], [33]]. Karnataka state had better contact tracing in the country during the early stage of the pandemic by detecting 47·4 contacts per confirmed COVID-19 case [19]. Hence, we believe that there was no bias in selection of cases which could have skewed the observation that more than 90% of cases were asymptomatic. Indeed, it is merely a reflection of the COVID-19 scenario during the early stage of pandemic in the state of Karnataka. Although, it may be argued that some of the asymptomatic cases may have developed symptoms after 7–13 days, all the cases during the lockdown period were hospitalized and were under medical observation for 14 days. Therefore the chances of misclassification was unlikely [28,34]. Moreover, the selection of contacts or cases in the testing system was according to the national guidelines that mandated active search and testing for SARS-CoV-2 infection in each state [20].

There was no risk factor ascribed for acquisition of SARS-CoV-2 in 267/3404 cases (7·8%) in the present study. There are two likely explanations for this. Firstly, it may be argued that it is indicative of “community transmission”. Recent reports of SARS-CoV-2 seroprevalence studies carried out across India have indicated that substantial transmission has occurred in rural areas although it was higher in urban settings. Further, it was noted that a mere 3% of seropositive individuals reported COVID-19 symptoms [35,36]. Secondly, owing to limitations in routine surveillance data collection, especially during the early phase of the epidemic in India as reported by an earlier study, it is likely that the 267 cases are indicative of community transmission [37].

There were two limitations to this study. Firstly, the study was restricted to the early period of the pandemic when large number of infections occurred due to travel. As a result, complete contact tracing data was available only in 966/3404 (28·4%) COVID-19 cases until May 31, 2020. Although contact tracing for all cases would be ideal, given the reality of the rapid and extensive spread of SARS-CoV-2 virus, there were numerous constraints and challenges in achieving this. Secondly, a key factor in the transmission of COVID-19 is the high level of shedding of SARS-CoV-2 virus from the upper respiratory tract. Viral RNA shedding is higher at the time of symptom onset and declines after days or weeks [38]. The difficulty of distinguishing asymptomatic cases from those who are pre-symptomatic is a major stumbling block. Since longitudinal data on the occurrence of symptoms after the date of collection of specimens was available for the quarantine period (14 days) of this study, it is plausible that a few cases with extreme incubation periods may have become symptomatic [34]. Consequently, we relied upon available credible data to understand the dynamics of transmission of SARS-CoV-2 and firmly believe the results of this study covering nearly 3500 positive cases (by RT PCR), from an Indian state with a population size of more than 65 million offers significant lessons for most countries of the Global South.

In conclusion, the findings of this study demonstrate that, during the early phase of the pandemic, spread of SARS-CoV-2 infection in the state of Karnataka occurred from both asymptomatic and symptomatic individuals. While preventive measures such as use of face masks, social distancing, isolation, quarantine and hand washing are essential to curtail the spread of SARS-CoV-2 from both asymptomatic and symptomatic individuals, public health policy should emphasize on the higher potential of symptomatic cases to transmit the infection to a larger number of contacts during the early phase of pandemic. Active search and early detection of symptomatic elderly patients with comorbidities should be prioritized to reduce mortality. Further, surveillance and testing of all individuals in high-risk settings to capture ‘silent’ transmission by asymptomatic individuals should be an integral component of the policy to prevent spread of SARS-CoV-2.

Data sharing statement

Study data (in spreadsheet form, with included data definitions) is available with the Office of the State Surveillance Officer, Government of Karnataka. Data sharing requests are subject to approval by the Commissioner of Health and Family Welfare, Government of Karnataka. Data and other information can be requested by sending an email to the corresponding author at: virusravi@gmail.com.

Author contributions

NK and SKSH contributed equally to data collection, analysis and preparation of manuscript, GRB contributed to analysis and review of manuscript, MMV contributed to laboratory testing, data analysis and review of the manuscript, PD and PKBG contributed to collection, collation of data and preparation of line-list, DAJ contributed to data analysis and figures, AD contributed to data analysis, preparation and review of the manuscript and VR conceptualized the study and contributed to data analysis, preparation and review of the manuscript.

Declaration of Competing Interest

The authors declare no conflict of interest.