ObjectivesOur primary objective is to predict the dynamics of COVID-19 epidemic in India while adjusting for the effects of various progressively implemented containment measures. Apart from forecasting the major turning points and parameters associated with the epidemic, we intend to provide an epidemiological assessment of the impact of these containment measures in India. MethodsWe propose a method based on time-series SIR model to estimate time-dependent modifiers for transmission rate of the infection. These modifiers are used in state-space SIR model to estimate reproduction number R 0 , expected total incidence, and to forecast the daily prevalence till the end of the epidemic. We consider four different scenarios, two based on current developments and two based on hypothetical situations for the purpose of comparison. ResultsAssuming gradual relaxation in lockdown post 17 May 2020, we expect the prevalence of infecteds to cross 9 million, with at least 1 million severe cases, around the end of October 2020. For the same case, estimates of R 0 for the phases no-intervention, partial-lockdown and lockdown are 4.46 (7.1), 1.47 (2.33), and 0.817 (1.29) respectively, assuming 14-day (24-day) infectious period. ConclusionsEstimated modifiers give consistent estimates of unadjusted R 0 across different scenarios, demonstrating precision. Results corroborate the effectiveness of lockdown measures in substantially reducing R 0 . Also, predictions are highly sensitive towards estimate of infectious period.

In the absence of sufficient testing capacity for COVID-19, a substantial number of infecteds are expected to remain undetected. Since the undetected cases are not quarantined, they can be expected to transmit the infection at a much higher rate than their quarantined counterparts. That is, in the absence of extensive random testing, the actual prevalence and incidence of the SARS-CoV-2 infection can be significantly higher than that being reported. Thus, it is imperative that the information on the percentage of undetected (or unreported) cases be incorporated in the mechanism for estimating the key epidemiological parameters, like rate of transmission, rate of recovery, reproduction rate, etc. , and hence, for forecasting the transmission dynamics of the epidemic. In this paper, we have developed a new dynamic version of the basic susceptible-infected-removed (SIR) compartmental model, called the susceptible-infected (quarantined/ free) - recovered- deceased [SI(Q/F)RD] model, to assimilate the impact of the time-varying proportion of undetected cases on the transmission dynamics of the epidemic. Further, we have presented a Dirichlet-Beta state-space formulation of the SI(Q/F)RD model for the estimation of its parameters using posterior realizations from Gibbs sampling procedure. As a demonstration, the proposed methodology is implemented to forecast the COVID-19 transmission in California and Florida. Results suggest significant amount of underreporting of cases in both states. Further, posterior estimates obtained from the state-space SI(Q/F)RD model show that average reproduction numbers associated with undetected infectives [California: 1.464; Florida: 1.612] are substantially higher than those associated with quarantined infectives [California: 0.497; Florida: 0.359]. The long-term forecasts show similar trend as that of the estimates of excess deaths for the comparison period post training data timeline.

Estimation of Quality Adjusted Life Years (QALYs) is pivotal towards economic evaluation and cost-effectiveness analysis of medical interventions. Most of the methods developed till date for calculating QALYs are based on multi-state structures where fixed utility values are assigned to each disease state and total QALYs are calculated on the basis of total lengths of stay in each state. In this article, we have presented a new proxy approach to define utility as a function of risk factors, which can be used to calculate QALY without defining discrete disease states. Retrospective survival data of HIV/ AIDS patients undergoing treatment at the Antiretroviral Therapy (ART) center of Ram Manohar Lohia hospital in New Delhi has been used to demonstrate implementation of the proposed methodology. Joint modelling, with a mixed effect longitudinal sub-model for CD4 count and a Cox proportional hazard survival sub-model with time dependent covariates, has been used to estimate risks associated with different factors and covariates. Using the proxy utilities, QALYs have been calculated for each individual for their lifetime time horizon, defined as the time since their registration in the ART till death or till their age reach average life expectancy of HIV/ AIDS patients in India. QALY results are consistent with findings of conventional cost-effectiveness studies on ART for HIV/ AIDS patients in India.

In the absence of sufficient testing capacity for COVID-19, a substantial number of infecteds are expected to remain undetected. Since the undetected cases are not quarantined, they are expected to transmit the infection at a much higher rate than their quarantined counterparts. That is, under the lack of extensive random testing, the actual prevalence and incidence of the SARS-CoV-2 infection may be entirely different from that being reported. Thus, it is imperative that the information on the percentage of undetected (or unreported) cases be considered while estimating the parameters and forecasting the transmission dynamics of the epidemic.In this paper, we have developed a new version of the basic susceptible-infected-removed (SIR) compartmental model, called the susceptible-infected (quarantined/ free) -recovered-deceased [SI(Q/F)RD] model, to incorporate the impact of undetected cases on the transmission dynamics of the epidemic. Further, we have presented a Dirichlet-Beta state-space formulation of the SI(Q/F)RD model for the estimation of its parameters using posterior realizations from Gibbs sampling procedure. As a demonstration, the proposed methodology is implemented to forecast the COVID-19 transmission in California and Florida.HighlightsData calibrated for underreporting using excess deaths and case fatality rate.A new extension of SIR compartmental model, called SI(Q/F)RD, is introduced.A Dirichlet-Beta state-space formulation of the SI(Q/F)RD model is developed.Gibbs sampling used to estimate the Bayesian hierarchical state-space model.Proposed methodology is applied on the COVID-19 data of California and Florida.

In the absence of sufficient testing capacity for COVID-19, a substantial number of infecteds are expected to remain undetected, and hence, are not quarantined. This, in turn, defeats the whole purpose of non-medical containment measures, like quarantine, lockdown, travel ban, physical distancing etc., by keeping the average reproduction rate above 1. To stress upon the importance of extensive random testing for breaking the chains of transmissions, we have formulated a detailed framework for carrying out cost-effectiveness analysis (CEA) of extensive random testing in comparison to targeted testing (the existing testing policy followed by most countries). This framework can be easily extended for CEA of any other non-medical or even medical interventions for containing epidemics.We have developed a new version of the basic susceptible-infected-removed (SIR) compartmental model, called the susceptible-infected (quarantined/ free) - recovered-deceased [SI(Q/F)RD] model, to incorporate the impact of undetected cases on the transmission dynamics of the epidemic. Further, we have presented a Dirichlet-Beta state-space formulation of the SI(Q/F)RD model for the estimation of its parameters using posterior realizations from Gibbs sampling procedure. As an application, the proposed methodology is implemented to forecast the COVID-19 transmission in California and Florida, and further carry out CEA of extensive random testing over targeted testing.HighlightsEstimated values of excess deaths associated with COVID-19 are used to account for underreporting, and for calibrating data to obtain actual counts of cases.A new flexible version of SIR compartmental model, called SI(Q/F)RD, is introduced to facilitate in the CEA exercise.Dirichlet-Beta state-space formulation of the SI(Q/F)RD model is used to predict the transmission dynamics of the epidemic.CEA is conducted in terms of outcome (reduction in infections and deaths) and total cost of tests.Proposed methodology is applied on the data of California and Florida.