Application of Predictive Models to Assess the Association Between Immunoprophylaxis Indicators and Measles Incidence

Introduction: Humanly Possible: Immunization for All – that was the theme of World Immunization Week 2024. Vaccination is one of the most successful public health interventions of all time, which has prevented significant health losses. This article provides a rationale for applying machine learning in the analysis of complex spatiotemporal relationships between vaccination coverage and measles incidence.

Objective: To analyze the impact of different vaccination factors on measles incidence using predictive models.

Materials and Methods: Panel data covering 85 regions of the Russian Federation from 1996 to 2022 were used in the study (over 2,200 observations). We applied cross-correlation, ordinary least squares (OLS) regression, ARIMA, and CatBoost gradient boosting models. SHAP analysis was used for model interpretation.

Results: Cross-correlation analysis showed negative associations between current vaccination levels and incidence, as well as a delayed association between revaccination and incidence with a three-year lag. The OLS model demonstrated pronounced overfitting under conditions of multicollinearity among lagged predictors. The ARIMA model showed unsatisfactory predictive performance on the test set. On the independent test set, the CatBoost model demonstrated satisfactory predictive performance (R² = 0.529) and outperformed OLS and ARIMA in terms of generalization ability. According to the SHAP analysis, a three-year lag in revaccination and a one-year lag in adult vaccination coverage made the largest contribution to model predictions.

Conclusion: Application of CatBoost combined with SHAP analysis made it possible to identify nonlinear relationships between vaccination indicators and measles incidence and can be used for evaluation of regional immunization programs.

1. Salmon DA, Dudley MZ, Glanz JM, Omer SB. Vaccine hesitancy: Causes, consequences and a call to action. Am J Prev Med. 2015;49(6 Suppl 4):S391-S398. doi: 10.1016/j.amepre.2015.06.009

2. Phadke VK, Bednarczyk RA, Salmon DA, Omer SB. Association between vaccine refusal and vaccine-preventable diseases in the United States: A review of measles and pertussis. JAMA. 2016;315(11):1149-1158. doi: 10.1001/jama.2016.1353

3. Bi K, Nguyen T, Peng B, et al. Modeling the impact of MMR vaccination strategies on measles outbreaks in Texas. JAMA Health Forum. 2025;6(9):e253992. doi: 10.1001/jamahealthforum.2025.3992

4. Hviid A, Vinsløv Hansen J, Frisch M, Melbye M. Measles, mumps, rubella vaccination and autism: A nationwide cohort study. Ann Intern Med. 2019;170(8):513-520. doi: 10.7326/M18-2101

5. Martinelli D, Tafuri S, Fortunato F, Cozza V, Germinario CA, Prato R. Are we ready to abrogate compulsory vaccinations for children? Hum Vaccin Immunother. 2015;11(1):146-149. doi: 10.4161/hv.34417

6. Bogacheva NV, Ananina EA. Analysis of measles incidence and causes in present-day conditions. Meditsinskoe Obrazovanie Segodnya. 2024;(1(25)):38-42. (In Russ.)

7. Kaunina DV, Vasilieva TP, Russkikh SV. Results of validating the proprietary method of establishing personal levels of vaccination adherence. Zdorov’e Naseleniya i Sreda Obitaniya. 2023;31(8):17-28. (In Russ.) doi: 10.35627/2219-5238/2023-31-8-17-28

8. Tsvirkun OV, Tikhonova NT, Turaeva NV, Ezhlova EB, Melnikova AA, Gerasimova AG. Population immunity and structure of measles cases in the Russian Federation. Epidemiologiya i Vaktsinoprofilaktika. 2020;19(4):6-13. (In Russ.) doi: 10.31631/2073-3046-2020-19-4-6-13

9. Kaunina DV. Determinants of infectious diseases immuneprophylaxis organization. Byulleten’ Natsional’nogo Nauchno-Issledovatel’skogo Instituta Obshchestvennogo Zdorov’ya Imeni N.A. Semashko. 2023;(1):17-23. (In Russ.) doi: 10.25742/NRIPH.2023.01.003

10. Agapova TE, Mulendeeva ES, Anisimova TA. Measles in Russia: Epidemiological trends 2010–2024. Biologiya i Integrativnaya Meditsina. 2025;(3(75)):139-151. (In Russ.) doi: 10.24412/cl-34438-2025-139-151

11. Plakida AV, Briko NI, Namazova-Baranova LS, Feldblyum IV, Los’ NA, Ivanova ES. Increasing population adherence to vaccination: Evaluation and a systematic approach to implementation. Epidemiologiya i Vaktsinoprofilaktika. 2022;21(3):4-26. (In Russ.) doi: 10.31631/2073-3046-2022-21-3-4-26

12. Golubkova AA, Platonova TA, Smirnova SS, Kovyazina SA. The epidemic process of measles with different strategies of vaccination in a large industrial center of the Middle Urals. Epidemiologiya i Vaktsinoprofilaktika. 2019;18(1):34-42. (In Russ.) doi: 10.31631/2073-3046-2019-18-1-34-42

13. Simakova YaV, Gushchin VA, Semenenko TA, et al. Characteristics of the epidemic process of measles, rubella and mumps in Moscow and assessment of their seroprevalence at the present stage. Voprosy Virusologii. 2025;70(2):133-146. (In Russ.) doi: 10.36233/0507-4088-282

14. Semenenko TA, Nozdracheva AV, Asatryan MN, Akimkin VG. Multivariate analysis of the megacity population immunity to measles. Vestnik Rossiyskoy Akademii Meditsinskikh Nauk. 2019;74(5):351-360. (In Russ.) doi:10.15690/vramn1170

15. Nozdracheva AV, Semenenko TA, Asatryan MN, et al. Immunological susceptibility of metropolis population to measles in its elimination stage. Epidemiologiya i Vaktsinoprofilaktika. 2019;18(2):18-26. (In Russ.) doi: 10.31631/2073-3046-2019-18-2-18-26

16. De Serres G, Gay NJ, Farrington CP. Epidemiology of transmissible diseases after elimination. Am J Epidemiol. 2000;151(11):1039-1052. doi: 10.1093/oxfordjournals.aje.a010145

17. Romanenko TA, Horokhova AA. Monitoring of preventive immunization coverage of population during the elimination phase of measles and rubella. Vestnik Gigieny i Epidemiologii. 2020;24(4):395-399. (In Russ.)

18. Briko NI, Feldblum IV, Alyeva MKh, et al. The concept of risk communications to ensure adherence to vaccination as a necessary component of the strategic development of immunoprophylaxis in Russia. Obshchestvennoe Zdorov’e. 2021;1(1):32-43. (In Russ.) doi: 10.21045/2782-1676-2021-1-1-32-43

19. Nozdracheva AV, Semenenko TA. The status of herd immunity to measles in Russia: A systematic review and meta-analysis of epidemiological studies. Zhurnal Mikrobiologii, Epidemiologii i Immunobiologii. 2020;97(5):445-457. (In Russ.) doi: 10.36233/0372-9311-2020-97-5-7

20. Zverev VV, Yuminova NV. Vaccines. Prevention of viral infections from e. Jenner to date. Voprosy Virusologii. 2012;(S1):33-42. (In Russ.)

21. Ljashenko VA, Juminova NV, Aleksander SK. Some variants of measles vaccine: Experimental researches. Epidemiologiya i Vaktsinoprofilaktika. 2006;(3(28)):19-23. (In Russ.)

22. Ivanova RP. Correlation analysis of innovation processes in light industry. Vestnik Nizhegorodskogo Universiteta im. N.I. Lobachevskogo. Seriya: Sotsial’nye Nauki. 2016;(4(44)):27-34. (In Russ.)

23. Fashutdinova VA. [Joint analysis of time series of geo-seismic activity and geocentre locations.] Vestnik Ul’yanovskogo Gosudarstvennogo Tekhnicheskogo Universiteta. 2009;(3(47)):16-20. (In Russ.)

24. Prokhorenkova L, Gusev G, Vorobev A, Dorogush AV, Gulin A. Catboost: Unbiased boosting with categorical features. In: Proceedings of the 32nd Conference on Neural Information Processing Systems (NeurIPS), Montreal, QC, December 2–8, 2018. Montreal, Canada; 2018:6638-6648. Accessed April 15, 2026. https://proceedings.neurips.cc/paper_files/paper/2018/file/14491b756b3a51daac41c24863285549-Paper.pdf