Activity-based planning of primary care utilization in a model of prospective demand for workforce Authors: Ilja Stepanov, Igor Sheiman Published: 2025 Introduction A shortage of primary health care (PHC) workers has become an increasingly significant health issue in many countries [1], including Russia [2]. Despite holding a leading position globally in terms of the number of physicians per capita, the country is currently experiencing an acute deficit of healthcare professionals. Studies indicate that this is a structural problem: a shortage of physicians in outpatient settings coexists with a surplus in hospitals; some specialties face shortages, while others exceed demand; the limited functions of medical nurses create additional demand for doctors; and there is substantial regional inequality [2, 3]. These imbalances are largely attributed to shortcomings in the planning system. While Russia has established a system for current health workforce planning, it lacks strategic planning for the medium and long term [3]. The government has set a goal to create such a system, and an expert group has been established to develop a methodology for strategic planning. In the literature, four types of strategic planning models are distinguished: health workforce-to-population ratio planning, utilization-based planning, needs-based planning, and target-based planning. Needs-based models are promising as they account for patient-oriented medical care expectations, but these expectations are challenging to evaluate and predict [4]. The Russian system of current workforce planning is based on indicators of care volumes, while a long-term planning system is still being developed. Therefore, it is unrealistic to begin with complex, multi-factor needs-based models. Instead, a utilization-based model of strategic planning has been proposed for Russia by the expert group. This model is based on the triad of: morbidity, utilization of health care, and labor per unit of care utilization (staffing norms). This paper focuses on forecasting future needs in PHC utilization, specifically the number of physician visits per capita. In addition to the traditional approach of determining care utilization based on general factors (mostly demographic), an original approach is suggested that assesses the impact of expected policy activities aimed at strengthening primary care. Key activities are identified, including those that potentially reduce the need for physicians. The objective of this paper is to present a new model for strategic planning of the need for primary care workers in Russia, as well as its most critical element—a methodology for anticipating the number of physician visits, and therefore the number of primary care physicians needed to serve these visits. Within the framework of strategic planning, a model for the supply of health workers, accounting for their influx and exit, has also been developed. A demonstration evaluation has been conducted; however, due to the limited scope of the case study, only the activity-based approach to evaluating PHC utilization for determining demand for primary care physicians is presented. Basic elements of the model for prospective demand for PHC workforce The demand model is estimated for the entire country and its regions based on a new integrated forecasting model. The model is focused on estimating the number of primary care physicians that includes generalists of a so called “district service” (district therapists, district pediatricians, general practitioners) and outpatient specialists. Contrary to post-communist Central and Eastern European countries, Russian primary care is based on highly consolidated service provision in big multispecialty facilities and the dominance of public ownership. General practice is still in its infancy, while an attempt has been made to compensate for this with a growing number of specialists in primary-care settings. The overspecialization is a specific characteristic of primary care organization in Russia [5, 6]. This characteristic was taken into account in long-term planning. Demand forecast consists of the following elements (successive steps). Determining the projected population size and morbidity In the model, official government demographic forecasts are used, and morbidity is estimated by disease class and extrapolated to the forecast period. Assumptions are made regarding deviations in morbidity dynamics for individual classes, either upward or downward. Aggregated population and morbidity data are used without breakdowns by age and sex groups, which may reduce the accuracy of projected parameters but facilitates the assessment of the macro impact of individual policy activities on PHC utilization and workforce demand. Further development and validation of the approach will determine the necessary level of data disaggregation to be used in subsequent modeling initiative. Assessing aggregate PHC utilization A major parameter—the prospective number of total physician visits per capita per year—is determined by assessing the impact of general factors such as population and morbidity dynamics, the emergence of new types of medical organizations, major technological changes, shifts in the ratio of outpatient and inpatient care, and health expenditure trends. The utilization vector is determined for each factor on a scenario basis. The future aggregate utilization under the most realistic scenario serves as a control value for subsequent calculations of the impact of individual policy activities. If the sum of their impacts significantly deviates from the aggregate value, an analysis is conducted to identify the reasons for the discrepancy, and adjustments are made to the disaggregated estimates. A preliminary assessment suggests that factors increasing the volume of primary health care in the medium term will dominate over those reducing it. Under the preferred scenario, the need for physician visits per capita increases from 7.9 in 2023 to 9.0 in 2030. In the subsequent period, up to 2036, this need will gradually decrease to 8.8 visits. Evaluating the activity-based number of physician visits A set of policy activities is identified based on priorities reflecting the most significant problems in primary care, as presented in previous studies [5, 6]. Two groups of activities are identified: Group A, which increases demand for physician visits, and Group B, which decreases demand. These activities are adjusted for 2036 (Table 1). Table 1Policy activities to strengthen PHC that are accounted for planning physician visits per capita in 2030Group A. Activities increasing the need for physician visitsGroup B. Activities decreasing the need for physician visitsDeveloping the capacity of PHC institutions (A1)Developing general practice to decrease the workload of outpatient specialists (B1)Improving the effectiveness of disease prevention (A2)Delegating some physician functions to medical nurses (B2)Expanding and deepening dispensary management of patients with chronic diseases and comorbidity (A3)Increasing access to outpatient medicines(B3)Increasing the volumes of rehabilitation and palliative care provided by PHC institutions (A4)Developing remote monitoring of patient health status (B4)Enhancing the capacity of primary care physicians to take on a part of the workload of inpatient care providers (A5)Increasing continuity of care after hospital discharge of complex patients (A6) Policy activities to strengthen PHC that are accounted for planning physician visits per capita in 2030 Increasing access to outpatient medicines (B3) Evaluating physician visits across specialties The total number of visits is detailed for individual physician specialties using coefficients for each type of specialist. These coefficients indicate deviations in the number of visits per capita relative to the average for all specialties. Values are determined using a regression model that measures the dependency between the number of visits per capita for a specific specialty (dependent variable) and morbidity by disease classes (predictors) over 2012–2023. Specialists' involvement in treating other diseases is also considered. For example, the number of visits to obstetricians and gynecologists is estimated using a regression model that accounts for diseases in multiple classes, such as “pregnancy, childbirth, and postpartum period,” “neoplasms,” “skin diseases,” and “diseases of the genitourinary system.” This approach results in higher coefficients for certain specialties. A special approach is used for district service physicians, who have a catchment population (patient list). Targets for this indicator are set, aiming to reduce the number of people served by one district physician by 10% in 2030 and by 20% in 2036 compared to 2023 (1877 people per district physician). The total number of primary care physicians is the sum of outpatient specialists and district physicians. Assessing labor per unit of primary care utilization In Russia, standards (“normatives”) exist for the expected number of visits per physician per year for individual categories of specialists. These standards are close to the staffing norms used internationally [7]. They are adjusted by an expert group, considering factors such as: Predicted dynamics of morbidity for individual disease classes. Predicted dynamics of morbidity for individual disease classes. Shifting care from inpatient to outpatient settings. Shifting care from inpatient to outpatient settings. Delegating some physician functions to nurses. Delegating some physician functions to nurses. Regional estimates account for unique territorial characteristics, such as population density, which affects physician workload. National standards are adjusted using coefficients determined by the Federal Ministry of Health. Determining the number of medical nurses The predicted ratio between physicians and nurses is used to estimate the number of nurses. Changes in this ratio result from: (a) Delegating some physician functions to nurses [8], (b) New factors driving demand for nurses [9]. In 2023, the ratio of nurses to physicians in PHC in Russia was 1.58:1 [10]. The average ratio for 37 OECD countries is 2.5:1, while post-Soviet countries average 2.2:1 [11]. The preferred scenario aims to reach a ratio of 2.1:1 by 2036 and 1.8:1 by 2030. Adjustment of estimates Adjustments are made to account for: Overestimation of the impact of certain activities. Overestimation of the impact of certain activities. Significant gaps between estimated workforce needs and expected supply. Significant gaps between estimated workforce needs and expected supply. Rejection of estimates by health authorities as unrealistic or inappropriate. Rejection of estimates by health authorities as unrealistic or inappropriate. The impact of PHC policy activities on the number of physician visits It includes a qualitative and quantitative assessment of the impact of each activity to strengthen PHC and a quantitative assessment of the aggregate impact of all activities, as indicated in Fig. 1. Methodological scheme for determining the degree of the impact of activities in group A and group B on the number of physician visits per capita The formula for the aggregate impact for each of the 10 activities is as follows: Where: Nvforecastperpersonperyer—prospective number of physician visits per capita; Nvbaseyear(2023)perpersonperyer—number of physician visits per capita in base period—7.9 visits in 2023; I—aggregate coefficient of the impact of all activities. The aggregate coefficient of the impact of all activities is measured according to the formula: where: kAj—values of coefficients kA1-kA7, measuring a degree of the positive impact of group A activities on the number of physician visits per capita in a prospective period (kAj>1). kBj—values of coefficients kB1-kB4, measuring a degree of the negative impact of group B activities on the number of physician visits per capita in a prospective period (kBj<1). Coefficients kAj and kBj are assessed based on the following factors FAji, FBji that explain the impact of the activities of both groups according to the formulas: where: FAji, FBji—indicators of visits per capita growth that correspond to the activities in the prospective period (Nvforecastperpersonperyer). For each coefficient kAj or kBj, one or more indicators of FAji, FBjiare used,where i—index number. For each indicator FAji, FBji,a weight is attached WtAji or WtBji Weight values are used in the range from 0 to 100%. The weight of the indicator shows the extent to which it explains the value of the coefficient kAj or kBj. For example, the increase in the capacity of polyclinics (the indicator of the activity A1) will explain the impact on the predicted number of visits per capita to the extent that corresponds to the share of visits to polyclinics in the total volume of visits to all types of PHC institutions (polyclinics, physician clinics, mobile PHC stations). The assessment is made with the use of quantitative and expert methods. The quantitative method is used to assess the expected growth of visits based on the statistical data or standards (normatives) of visits per type of service per capita. These standards are set by the government in the annual federal program of state guarantees of free care provision, a major document of planning service delivery (PSG). The expert method involves assessing the current and future situation in primary care and studying international developments and best practices from Russian regions [3, 5]. The evaluation procedure for each activity is presented below. Activity A1. Developing the capacity of PHC institutions The government has planned the construction and additional equipment needed for primary health care institutions for the period up to 2030. Assumptions are made about the increase in the capacity of three types of primary care institutions. The values of FA11 – FA13 are determined for 2030: the increase in the number of physician visits as a result of the increase in the capacity of polyclinics will be 10% (FA12), physician clinics—10% (FA12), mobile primary health care stations—5% (FA13). The weight of each indicator corresponds to the share of visits to these three types of institutions in the base year: 70% (WtA11), physician clinics—25% (WtA12), mobile primary care stations—5% (WtA13). According to formula 3, the value of the coefficient of the activity A1 impact is determined: Activity A2. Improving the effectiveness of disease prevention The country has introduced a vertical large-scale program of ‘dispensarization’ that includes health check-ups and screenings for the entire adult population. It is expected to improve the results of medical interventions and ensure health gains at a relatively low cost [12]. Its coverage and scope is increasing for people who have had Covid-19. A new version of the program (in-depth dispensarization) is based on the utilization target of the number of preventive episodes—bundled physician preventive visits per resident. An episode may include two and more visits. The normative value of the number of episodes is set by the PSG—0.051 per capita in 2023 [13]. We assume that each episode consists of two actual physician visits in the base period. We also assume that this number will increase by 80% in 2030. The impact of activity A2 on the total number of physician visits is determined, taking into account the weight of the visit growth indicator. This weight corresponds to the share of visits for in-depth dispensarization in the total number of physician visits in the base year—1.29% (WtA21). The value of the coefficient of the impact of the activity A2 is determined by formula 3: Activity A3. Expanding and deepening dispensary management of patients with chronic diseases Dispensary management in Russia is similar to chronic disease management programs, common internationally. It is planned by the PSG as a number of episodes per capita—0.262 in the base year [14]. We assume that each episode consists of three physician visits in the base period. We also assume that this number will increase by 80% in 2030 (FA31) due to the growing priority of this activity. The impact of the activity A3 on the total number of physician visits is determined, taking into account the weight of the visit growth indicator. This weight corresponds to the share of visits for in-depth dispensary management of chronic cases in the total number of physician visits in the base year—9.9% (WtA31). The value of the coefficient of the impact of the activity 3 is determined using formula 3: Activity A4. Increasing the volumes of rehabilitation and palliative care provided by PHC institutions The PSG sets the following normatives: the number of rehabilitation episodes per resident in 2023—0.002945, with 5 physician visits per episode; the number of rehabilitation episodes per resident in 2023—0.002945, with 5 physician visits per episode; the number of physician visits for provision of palliative care in 2023—0.03 per capita [14]. the number of physician visits for provision of palliative care in 2023—0.03 per capita [14]. These types of care are at the initial stage of their development in Russia, therefore it was assumed that the number of respective visits (FA41 и FA42) will increase by 100% in 2030. The impact of the activity A4 on the total volume of physician visits is determined taking into account the weight of the visit growth indicator. The weight for rehabilitation care corresponds to the share of such visits in the total volume of physician visits in the base year—0.19% (WtA41), for palliative care—0.38% (WtA42). The value of the impact coefficient of the activity A4 is determined using formula 3: Activity A5. Enhancing the capacity of primary care physicians to take on a part of the workload of inpatient care providers In Russia, the hospital-centric model of care still prevails [3], therefore it is planned to shift some care to primary care settings. Preliminary assumption: managing patients with some diseases of the digestive and respiratory organs. To assess the impact of the activity A5, an assumption is made about the increase in physician visits per capita by 3% (FA51) in 2030—as a result of replacing the corresponding volume of inpatient care. The impact of this activity on the total volume of physician visits is determined taking into account the weight of the visit growth indicator. The following estimates are used to determine this weight: the incidence of the selected diseases in 2023—0.57 cases per capita; the incidence of the selected diseases in 2023—0.57 cases per capita; the share of outpatient treatment of patients with these diagnoses in the total number of all patients with the corresponding diagnoses—70%; the share of outpatient treatment of patients with these diagnoses in the total number of all patients with the corresponding diagnoses—70%; the volume of outpatient care for patients with the selected diseases—3 physician visits per patient. the volume of outpatient care for patients with the selected diseases—3 physician visits per patient. The weight of these visits is 15.3% (WtA51). The value of the impact coefficient of the activity A5: Activity A6. Increasing continuity of care after hospital discharge of complex patients To assess the impact of this activity, it is assumed that the growth rate of complex patients after hospital admission, who are provided with outpatient close observation, will be 200% in 2030 relative to 2023. The following estimates are used to determine the weight of the visit growth indicator: 3 additional physician visits to provide care to these patients; the current level of post-discharge close observation is 3% of the total number of admissions per capita. Based on these estimates, it is calculated that the share of post-discharge physician visits in 2023 (corresponds to the weight of visit growth indicator) is 0.21% (WtA61). The value of the impact coefficient of the activity A6 (formula 3): Activity B1. Developing general practice A highly developed institute of GP can reduce the number of referrals to outpatient specialists. In Russia, the number of GPs is low—around 15% of district physicians [5]. Their development may contribute to a decrease in the total amount of physician visits. To assess the impact of this activity, it is assumed that in 2030 the negative growth rate of the number of visits to outpatient specialists per capita will be—5%. The share of visits to outpatient specialists in the total amount of physician visits is 30% in 2023 (it corresponds to the weight of the indicator of reducing these visits (WtB11). The value of the impact coefficient of the activity B1 (formula 3): Activity B2. Delegating some physician functions to medical nurses The previous research provides estimates of the specific physician functions that can be delegated to nurses with the subsequent decrease in the demand for physicians [8]. To assess the impact of this activity, it is assumed that in 2030 the negative growth rate of the number of physician visits per resident will be -3%. The value of the impact coefficient of the activity B2 (formula 3): Activity B3. Increasing the access to outpatient medicines Russia does not have health insurance for outpatient medicines. Only patients with selected diseases have this benefit under federal or regional programs. Increasing the access to free medicines, as well as development of electronic prescription, will contribute to a delayed effect of enhancing quality of care (particularly for chronic cases) with a subsequent decrease in demand for in-person physician visits. It is assumed that in 2030 the negative growth rate of the number of physician visits per resident will be – 2%. The value of the impact coefficient of the activity B3: Activity B4. Developing remote monitoring of patient health status In recent years, Russia has been intensifying activities to develop information technology in health, including remote monitoring of patients with chronic diseases using wearable medical devices. A significant increase in the volume of such services is expected. This is a factor of reducing demand for in-person physician visits. It is assumed that in 2030 the negative growth rate of the number of physician visits per patient with chronic disease will be—30%. To assess the weight of this indicator (WtB41), the share of visits for dispensary surveillance of chronic cases in the total amount of physician visits is used (9.9%). This is an estimate for the activity A3. The value of the impact coefficient of the activity B4: The aggregate impact of group A and group B policy activities The estimate of their aggregate impact on the number of physician visits per capita in 2030 using the coefficients is made according to the formula (1): where the aggregate coefficient of the impact of all activities is measured according to the formula (2): Thus, the activity-based estimate will be 8.74 physician visits per resident in 2030. It is slightly less than the estimate (9.0 visits in 2030) based on the evaluation of the aggregate trends in primary care. The adjusted number of visits is used in the following steps of the assessment of demand for PHC physicians. The preliminary estimate indicates that “Developing the capacity of PHC institutions” has the greatest impact on utilization and physician demand (kA1 = 1.0975), then goes “Expanding and deepening dispensary management of patients with chronic diseases” (kA3 = 1.0792). Sensitivity analysis The core of the model relies on a set of coefficients that quantify the impact of various policy activities on physician visit rates. These coefficients are based on expert judgment. To enhance the robustness of the model and make it more reproducible, we added a sensitivity analysis to measure changes in the aggregate coefficient of the impact of all activities (I) depending on variations in the values of the studied parameters. Changes in (I) are estimated using coefficients SAj,SBj (for each coefficient kAj and kBj) as shown in Table 2. The results determine the sensitivity coefficients (S) for the influencing factors under consideration. This analysis identifies which factors the proposed model is most and least sensitive to. Table 2The model of sensitivity analysisCoefficients SValues in the model (kAj-1), (kBj- 1)Changes ± 10% in (kAj-1), (kBj- 1)Changes in aggregate coefficient I with changes in (kAj-1), (kBj- 1) ± 10%Aggregate coefficient ISensitivity coefficients values SAj,SBj(1)(2)(3)(4)(5)(6) =|(4)/(5) - 1)/0.1|SA10.09750.10725/0.087750.11656/0.097060.106810.91SA20.010320.011352/0.0092880.107842/0.1057780.106810.10SA30.07920.08712/0.071280.11473/0.098890.106810.74SA40.00570.00627/0.005130.10738/0.106240.106810.05SA50.004590.005049/0.0041310.107269/0.1063510.106810.04SA60.00420.00462/0.003780.10723/0.106390.106810.04SB1− 0.015− 0.0165/− 0.01350.10531/0.108310.106810.14SB2− 0.03− 0.033/− 0.0270.10381/0.109810.106810.28SB3- 0.02− 0.022/− 0.0180.10481/0.108810.106810.19SB4- 0.0297− 0.03267/− 0.026730.10384/0.109780.106810.28 The model of sensitivity analysis Based on the distribution of sensitivity coefficient values (S), three groups of activities are distinguished: Group 1 (S ≥ 0.74), coefficients SA1, SA3. The model is most sensitive to the changes in PHC settings capacity (Activity A1) and in dispensary management of patients with chronic diseases and comorbidity (Activity A3). Group 2 (0.28 ≥ S ≥ 0.14), coefficients SB1—SB4. The model is moderately sensitive to developing general practice (Activity B1), delegating some physician functions to medical nurses (Activity B2), increasing access to outpatient medicines (Activity B3), developing remote monitoring of patient health status (Activity B4). Group 3 (S ≤ 0.10), coefficients SA2, SA4 – SA6. The model is least sensitive to the changes in the following factors: improving the effectiveness of disease prevention (Activity A2), increasing the volumes of rehabilitation and palliative care provided by PHC institutions (Activity A4), enhancing the capacity of primary care physicians to take on a part of the workload of inpatient care providers (Activity A5), increasing continuity of care after hospital discharge of complex patients (Activity A6.). By analyzing sensitivity coefficients, policymakers can focus resources and efforts on high-impact activities while ensuring that low-impact activities are appropriately managed without overestimating their importance. Discussion A proposed methodology is an adaptation of the traditional utilization-based model of health workforce demand. It seeks to partially overcome a limitation of the traditional model—its reliance on past and current utilization patterns, because it overlooks unmet needs or instances of overutilization of care. The proposed new model uses time series analysis of past trends to estimate prospective population and morbidity as an initial basis for prospective planning. The model seeks to address underutilization and overutilization for the base period, and the impact of current and planned policy activities to address these issues is estimated for future demand. This assessment is based on prior research on primary care developments in Russia and internationally [3, 5]. The model allows for approximating workforce demand estimates to align with the actual processes of current and future PHC development. Unlike routine econometric analysis, this approach determines potential causal links between policy activities and care utilization. In the Russian context, the implementation of the proposed model is facilitated by the government’s standardization of the number of visits per capita for certain types of care delivery. This standardization applies to policy activities in Group A (A1–A4). For these activities, the selected planning method is preferable, as it does not require statistical methods involving probabilistic assessments of impact factors. However, some activities are not standardized (e.g., developing general practice, delegating physician functions to nurses, increasing the availability of medicines in PHC) and therefore require the use of statistical tools. The assessment of demand for specific physician specialties is based on time series and regression analysis. Trends in morbidity for each class of diseases are used to predict service utilization, including for “related” classes of diseases. Even in this case, total and activity-based estimates of utilization are used as benchmarks. The sum of estimates for specific physician specialties is adjusted to align with the total utilization estimate. The potential of the proposed approach depends on the correct selection of priority activities that have the greatest impact on PHC volume and physician demand. These activities are determined by specific problems and features of primary care organization in the country, as well as current and expected priorities for its development. We selected activities planned by the government, along with additional activities dictated by long-term needs. The advantage of this approach lies in its relative simplicity for strategic workforce planning. It uses measurable indicators of physician visits per capita, which clearly explain the expected changes in PHC organization. The issue of uncertainty in long-term planning in this model [15] is addressed by developing expert assumptions about the parameters of policy activity impacts under various scenarios. These assumptions are quantified in terms of the impact of each activity on workforce needs. A preferred scenario is selected for each activity individually. The approach of simultaneously selecting scenarios for all activities was rejected due to its high complexity. Validation of the model is recommended with the involvement of external experts. The goal is to reach a consensus on the set of policy activities and the assumptions regarding the impact of each activity. Additionally, the compatibility of aggregate estimates with estimates of the impact of specific activities should be validated. A substantial gap between these estimates may indicate the need to adjust the assumptions. The major limitation of the model is its reliance on prior, arbitrary assumptions about the demand for medical nurses and other health workers. Using the ratio of nurses to physicians is a simplified approach that should eventually be replaced by direct evaluations of demand for each group of personnel using specific models. However, this limitation reflects the complexity of the roles and interdependence of various service provider groups, which are influenced by numerous factors and are difficult to predict. Simultaneously planning the demand for all workforce groups would greatly complicate the process. Therefore, we decided to start with evaluating physician demand as a foundation for building models for other health workforce groups with more detailed and empirically based assumptions. Additionally, some assumptions about the impact of certain activities were made without in-depth analysis. This limitation was partially addressed during the practical evaluation stage of the model. Some original assumptions were adjusted based on additional empirical data and input from new experts. It is worth noting that modeling and evaluation are separated in time, and gaps in empirical data are common at various planning stages. Similarly, the lack of age- and sex-specific morbidity data during the modeling stage can be compensated for by assessing the impact of individual policy activities on PHC utilization and workforce demand. Further development and validation of the approach will determine the degree of necessary data disaggregation. The major health policy recommendation is to link the strategic planning of health workforce to the selected policy activities to strengthen primary care. Conclusion The proposed model for strategic planning for the primary care workforce is based on care utilization models. In addition to the traditional approach of statistical analysis, this model focuses on assessing judgements from experts about the qualitative impact of current and expected policy activities. In the context of Russia, enhancing the capacity of PHC settings and expanding chronic disease management are the major drivers of increased demand for physician visits per capita and the number of primary care physicians. However, this model can also be useful for other countries implementing or planning medium- and long-term policy activities with substantial potential to increase or reduce workforce demand. The major strength of the model is its relative simplicity, while its primary weakness lies in its dependency on the correct selection of priority activities and assumptions about forecast parameters. Acknowledgements The article is an output of a research project implemented as part of the Basic Research Program at the National Research University Higher School of Economics. The authors are grateful to Dr. S. Shishkin, Dr. O. Obuhova, Dr. A. Bogomazova for the discussion of activity-based primary care utilization model and the joint work on the strategic workforce planning. Author contributions I.S. developed methodology of activity-based utlization evaluation, made statitical estiates, wrote the section, prepared Fig. 1 and Table 1 I. Sh. contributed to the methodology development, wrote and reviewed the main manuscript text, prepared Table 1. Funding The authors declare financial support was received for the research, authorship, and/or publication of this article. The paper was prepared in the framework of a research grant funded by the Ministry of Science and Higher Education of the Russian Federation. The authors also declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.