An Early Epidemiological Investigation of COVID-19 in Parsa,Nepal

Introduction

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection manifests as asymptomatic through to mild, severe, or fatal coronavirus disease (COVID-19). ¹ The effective reproduction number, R_t, determines the potential for epidemic spread and is an essential parameter to evaluate effectiveness of public health interventions in different settings. ² The basic reproduction number, R₀, is a special case which determines the potential for epidemic spread in a totally susceptible population at the beginning of an outbreak and is estimated to be between 2 and 4 for the wild-type SARS-CoV-2.

In this study, we conducted an early epidemiological investigation of individuals tested and confirmed as COVID-19 cases in Birgunj metropolitan city of Parsa district, and admitted to Narayani hospital, one of the busiest COVID-19-designated hospitals in Madhesh Province at the beginning of the COVID-19 pandemic.

Methods

The study was part of the regular COVID-19 response from the Epidemiology and Disease Control Division (EDCD) in Birgunj metropolitan city, Parsa district of Madhesh Province.

Epidemiological Information

Information for 12 COVID-19 cases were collected from March 30, 2020, to May 13, 2020, and the samples from contacts of the positive and suspected cases in quarantine centers were obtained from May 4 to 11, 2020.

After identification of the list of contacts, epidemiological information was collected following the World Health Organization (WHO) contact tracing and follow-up protocol.

Statistical Analysis

Attack rate (AR) and relative risks (RRs) were calculated with 95% confidence interval (CI) in different age groups. The positivity rate among the residents (representative of the ward population) of the wards of Birgunj metropolitan city were projected on the total ward population and adjusted by the reverse transcription polymerase chain reaction (RT-PCR) sensitivity and 95% CI estimated.

Reproduction numbers were calculated using either the AR for patient positive contacts (AR model) ³ or, when case dates were available, using both Poisson regression of the exponential growth (EG), defined by the per capita change in number of new cases per unit of time (EG model), ⁴ and a maximum likelihood (ML) approach (ML model), ⁵ after Obadia ⁶ using the R₀ package. For the EG and ML models, we used two generation intervals (GTs), as R₀ estimates are sensitive to this ⁷ : 3.95 and 5.2 days, with a gamma distribution and 1.5 standard deviation ⁷ (data not shown).

Results

SARS-CoV-2 Detection and Transmission in Parsa District

Of 662 samples collected from April 30 to May 13, 2020, 89 (13%, 95% CI [11%, 16%]) individuals tested positive for SARS-CoV-2. Among the total collected and tested, 478 samples were from Birgunj metropolitan city, of which 71 (15%, 95% CI [12%, 18%]) were SARS-CoV-2 positive. Moreover, 429 specimens were from inhabitants of Birgunj metropolitan city of which 66 (15%, 95% CI [12%, 19%]) tested positive for SARS-CoV-2.

The distribution of positive cases (Figure 1) through cross-sectional observation demonstrated the increasing pattern of COVID-19-positive cases with effective reproduction estimates ranging from 1.77 (95% CI [1.19, 2.6]) to 2.08 (95% CI [1.25, 3.45]) with the EG model and 3.95- or 5.2-day generation intervals, and 2.67 (95% CI [1.92, 3.61]) to 4.77 (95% CI [3.41, 6.43]) using the ML model (Figure 2).

OPEN IN VIEWER

Figure 1.

Trends of COVID-19 results from Parsa.

OPEN IN VIEWER

Figure 2.

Reproduction number of COVID-19.

Exposure Status of COVID-19 Cases

A small fraction of positive cases (14, 16%) had a travel history to COVID-19-affected countries and those cases were in facility-based quarantine. Only two (2%) of the returnees with SARS-CoV-2-positive results were not in quarantine. A source of acquiring infection was not determined for about 28% (25) of the confirmed COVID-19 cases.

When considering AR by the age group, our data revealed that most people who had prior contact with COVID-19 cases were from the working age groups, 15 to 59 years old (Table 1). However, children were found to have higher AR (AR = 47.4%, 95% CI [24%, 71%]). Adults (15-59 years) and elderly above 60 years of age had attack rates of 14.2% (95% CI [10%, 20%]) and 10.5% (95% CI [1%, 33%]), respectively. The risk of acquiring infection was not significantly different in female vs male, RR of 1.36 (95% CI [0.7, 2.58]). The transmission rate observed was significantly higher in people with a travel history to India than the people without travel history (RR = 2.4, 95% CI [1.01, 5.86]).

OPEN IN VIEWER

Table 1.

Age Structure of the Contacts and Attack Rate.

Age (years)	Number of SARS-CoV-2 positive	Number of contacts with the positive cases	Attack rate (%, 95% CI)
0-14	9	19	47.4 [24, 71]
15-59	31	218	14.2 [10, 20]
60 and above	2	19	10.5 [1, 33]

Abbreviation: SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Discussion

The AR in Birgunj metropolitan city was comparable with 0.50% to 59% of AR reported elsewhere at the beginning of the pandemic. ⁹ Our investigations of early cases in Birgunj metropolitan city revealed individuals had traveled to attend a large religious conference where transmission due to mixing people from different locations could occur, which is evident in other studies.

Our estimates of the reproduction number for the wild-type SARS-CoV-2 are similar to other studies. Our data were somewhat limited, and the different models suffer different issues, but our estimates are close or within the range reported previously, between 1.9 and 6.47 (median = 2.6, mean = 3.1). ¹⁰

Conclusion

Overall, our study confirmed that the COVID-19 outbreak in Nepal followed a similar pattern to that in other countries, with travelers introducing infection into the working-age populations and transmission through close contact. Our findings helped to strengthen public health and social measures at national level.