Risk factors and short-term projections for serotype-1 poliomyelitis incidence in Pakistan: A spatiotemporal analysis

Abstract

Background: Pakistan currently provides a substantial challenge to global polio eradication, having contributed to 73% of reported poliomyelitis in 2015 and 54% in 2016. A better understanding of the risk factors and movement patterns that contribute to poliovirus transmission across Pakistan would support evidence-based planning for mass vaccination campaigns.

Methods and findings: We fit mixed-effects logistic regression models to routine surveillance data recording the presence of poliomyelitis associated with wild-type 1 poliovirus in districts of Pakistan over 6-month intervals between 2010 to 2016. To accurately capture the force of infection (FOI) between districts, we compared 6 models of population movement (adjacency, gravity, radiation, radiation based on population density, radiation based on travel times, and mobile-phone based). We used the best-fitting model (based on the Akaike Information Criterion [AIC]) to produce 6-month forecasts of poliomyelitis incidence. The odds of observing poliomyelitis decreased with improved routine or supplementary (campaign) immunisation coverage (multivariable odds ratio [OR] = 0.75, 95% confidence interval [CI] 0.67-0.84; and OR = 0.75, 95% CI 0.66-0.85, respectively, for each 10% increase in coverage) and increased with a higher rate of reporting non-polio acute flaccid paralysis (AFP) (OR = 1.13, 95% CI 1.02-1.26 for a 1-unit increase in non-polio AFP per 100,000 persons aged <15 years). Estimated movement of poliovirus-infected individuals was associated with the incidence of poliomyelitis, with the radiation model of movement providing the best fit to the data. Six-month forecasts of poliomyelitis incidence by district for 2013-2016 showed good predictive ability (area under the curve range: 0.76-0.98). However, although the best-fitting movement model (radiation) was a significant determinant of poliomyelitis incidence, it did not improve the predictive ability of the multivariable model. Overall, in Pakistan the risk of polio cases was predicted to reduce between July-December 2016 and January-June 2017. The accuracy of the model may be limited by the small number of AFP cases in some districts.

Conclusions: Spatiotemporal variation in immunization performance and population movement patterns are important determinants of historical poliomyelitis incidence in Pakistan; however, movement dynamics were less influential in predicting future cases, at a time when the polio map is shrinking. Results from the regression models we present are being used to help plan vaccination campaigns and transit vaccination strategies in Pakistan.

Fig 1. Spatial distribution and trends in the incidence of poliomyelitis over time in different regions of Pakistan.

In (A), the spatial distribution of wild poliovirus type 1 (WPV1)-associated poliomyelitis cases in districts of Pakistan between January 2010 and December 2016 is shown (red dots). (B) Monthly confirmed WPV1-associated poliomyelitis cases in Pakistan reported between January 2010 and December 2016 are shown (bars). The same data are shown together with estimated serotype 1 vaccine-induced population immunity among children <36 months old (lines) for (C) Punjab, Sindh, Islamabad, Azad Jammu and Kashmir (AJK), and Gilgit-Baltistan, (D) Khyber Pakhtunkhwa, (E) Balochistan, and (F) the Federally Administered Tribal Area (FATA).

Fig 2. Spatial distribution of risk factors for wild poliovirus type 1 (WPV1)-associated poliomyelitis estimated from non-polio AFP data in districts of Pakistan for the period of July to December 2016.

(A) Vaccine-induced population immunity against serotype-1 poliomyelitis for children <36 months old. (B) Routine immunization (RI) cohort coverage. (C) Supplementary immunization activity (SIA) cohort coverage (values >100% indicate more SIA doses were reported than expected given the SIA calendar). Complete figures with earlier time periods are included in S1 Text (Figures B, C, and E).

Fig 3. Illustration of the estimated force of infection (FOI) resulting from the movement of infected individuals between districts during January to June 2014.

In (A), the components of the FOI are shown. Wild poliovirus type 1 (WPV1) cases in the previous 6 months (shown on the left) and estimated population movement calculated from the radiation model (shown for movement out of 2 chosen districts, centre, highlighted in dark blue) result in a district-specific FOI (right). The interplay between the FOI and the susceptibility of the population (population immunity, B) to determine the incidence of WPV1 cases in that 6-month period (C).

Fig 4. Reported and model-based estimates and forecasts of wild poliovirus type 1 (WPV1) cases between July 2013 and December 2016.

(A) Observed WPV1 cases. (B) Estimated probability of reporting at least 1 WPV1 case based on the best-fit regression model including all available data (January 2010–December 2016). Complete figures with earlier time periods included in Figures K and L in S1 Text. (C) Predicted probability of reporting at least 1 WPV1 case for the same periods using data up to the end of the preceding 6-month period. AUC, area under the curve.

Fig 5. Forecasts of wild poliovirus type 1 (WPV1) poliomyelitis and vaccination response for the period July 2016 through June 2017.

The estimated predicted probability of at least 1 WPV1 case for each district of Pakistan based on the best-fitting regression model is shown in (A) for July to December 2016 and (B) for January to June 2017. In (C), the planned supplementary immunization activity (SIA) calendar for January to June 2017 is shown based on national plans (as of January 2017).