Real-time analysis of the diphtheria outbreak in forcibly displaced Myanmar nationals in Bangladesh

Abstract

Background: Between August and December 2017, more than 625,000 Rohingya from Myanmar fled into Bangladesh, settling in informal makeshift camps in Cox's Bazar district and joining 212,000 Rohingya already present. In early November, a diphtheria outbreak hit the camps, with 440 reported cases during the first month. A rise in cases during early December led to a collaboration between teams from Médecins sans Frontières-who were running a provisional diphtheria treatment centre-and the London School of Hygiene and Tropical Medicine with the goal to use transmission dynamic models to forecast the potential scale of the outbreak and the resulting resource needs.

Methods: We first adjusted for delays between symptom onset and case presentation using the observed distribution of reporting delays from previously reported cases. We then fit a compartmental transmission model to the adjusted incidence stratified by age group and location. Model forecasts with a lead time of 2 weeks were issued on 12, 20, 26 and 30 December and communicated to decision-makers.

Results: The first forecast estimated that the outbreak would peak on 19 December in Balukhali camp with 303 (95% posterior predictive interval 122-599) cases and would continue to grow in Kutupalong camp, requiring a bed capacity of 316 (95% posterior predictive interval (PPI) 197-499). On 19 December, a total of 54 cases were reported, lower than forecasted. Subsequent forecasts were more accurate: on 20 December, we predicted a total of 912 cases (95% PPI 367-2183) and 136 (95% PPI 55-327) hospitalizations until the end of the year, with 616 cases actually reported during this period.

Conclusions: Real-time modelling enabled feedback of key information about the potential scale of the epidemic, resource needs and mechanisms of transmission to decision-makers at a time when this information was largely unknown. By 20 December, the model generated reliable forecasts and helped support decision-making on operational aspects of the outbreak response, such as hospital bed and staff needs, and with advocacy for control measures. Although modelling is only one component of the evidence base for decision-making in outbreak situations, suitable analysis and forecasting techniques can be used to gain insights into an ongoing outbreak.

Keywords: Bangladesh; Diphtheria; Epidemiological modelling; Health in humanitarian crises; Infectious disease; Mathematical modelling; Real-time modelling; Refugees.

Fig. 1

Outbreak analysis timeline with respect to the epidemic curve (black line). Green lines show the timing of events relevant to analysis: reporting of the first case, involvement of modellers at LSHTM, MSF decision on bed numbers required and MSF handover of the treatment centre. Blue lines show the date on which each of the four LSHTM forecasts was communicated to MSF

Fig. 2

Adjustment for the delay between symptom onset and case presentation (reporting delay). Evolution of the reporting delay (vertical axis) by week (horizontal axis) (a). Daily incidence of diphtheria cases in Balukhali (b) and Kutupalong (c) as reported within the first day after symptom onset (blue dots), adjusted for reporting delays (red dots) and as seen retrospectively (black line, data from 12 January 2018)

Fig. 3

Total incidence over all age groups and locations (a–d) and bed need as forecasted by the model. Black lines show data as reported by 12 January 2018, red dots the adjusted incidence and blue lines and shaded areas the median and 2.5% and 97.5% percentiles according to 1000 model runs forecasting from 12 December (a), 20 December (b), 26 December (c) and 30 December (d). Forecasts of bed need issued on the same dates (e). The horizontal line shows the number of beds provided as of a decision taken on 14 December

Fig. 4

Incidence by location (rows) and age group (columns) as forecasted by the model. Black lines show data as reported by 12 January 2018, and coloured lines and shaded areas the median and 2.5% and 97.5% percentiles according to 1000 model runs forecasting from 12 December (blue), 20 December (red), 26 December (purple) and 30 December (green)

Fig. 5

Posterior parameter values. Posterior ranges (vertical lines) and median values taken by model parameters for forecasts done on 12 December, 20 December, 26 December, 30 December and 8 January. The horizontal dashed lines show the mean value of the prior used for the proportion of reported. Uniform priors were used for other parameters