Enhancing the use of family planning service statistics using a Bayesian modelling approach to inform estimates of modern contraceptive use in low- and middle-income countries

Shauna Mooney, Leontine Alkema, Emily Sonneveldt, Kristin Bietsch, Jessica Williamson, Niamh Cahill · 2024

Read Enhancing the use of family planning service statistics using a Bayesian modelling approach to inform estimates of modern contraceptive use in low- and middle-income countries by Shauna Mooney, Leontine Alkema, Emily Sonneveldt, Kristin Bietsch, Jessica Williamson, Niamh Cahill (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-08606.txt.

Introduction

Shauna Mooney, Leontine Alkema, Emily Sonneveldt, Kristin Bietsch, Jessica Williamson, and Niamh Cahill add in a 2024 paper service statistics to a family planning estimation tool. Surveys every few years leave gaps in tracking modern contraceptive use. Clinic service counts are fresh and annual, yet noisy in country specific ways.

Their Bayesian route feeds yearly change in estimated modern use into the survey based tool. Uncertainty from each service data type is modeled rather than waved away. Extra country type uncertainty layers on top for shakier systems. Year to year change in the service signal predicts true change in use in many settings. The upgrade aims at tighter forecasts between survey rounds for program planners.

Catalog record

CREATORShauna Mooney, Leontine Alkema, Emily Sonneveldt, Kristin Bietsch, Jessica Williamson, Niamh Cahill
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE36 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-08606

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Enhancing the use of family planning service statistics using a Bayesian modelling approach to inform estimates of modern contraceptive use in low- and middle-income countries Shauna Mooney, Leontine Alkema, Emily Sonneveldt, Kristin Bietsch, Jessica Williamson, Niamh Cahill Abstract Background: Monitoring family planning indicators, such as modern contraceptive prevalence rate (mCPR), is essential for family planning programming. The Family Planning Estimation Tool (FPET) uses survey data to inform estimates and forecasts of family planning indicators, including mCPR, over time. However, reliance solely on surveys can lead to data gaps given that large-scale, population-health surveys are carried out on average every 3-5 years. Service statistics are a readily available data source that are routinely collected in conjunction with family planning service delivery. Various service statistics data types can be used to derive a family planning indicator called Estimated Modern Use (EMU). In a number of countries, annual rates of change in EMU have been found to be predictive of true rates of change in mCPR. However, it has been challenging to capture the varying levels of uncertainty associated with the EMU indicator across different countries and service statistics data types and to subsequently quantify this uncertainty when using EMU in FPET. Methods: In this paper, we present a new approach to using EMUs in FPET, to inform mCPR estimates. The new approach is based on using annual EMU rates of change as input, and accounts for uncertainty associated with the EMU derivation process. The approach also accounts for additional country-type-specific uncertainty. We assess the EMU type-specific uncertainty at the country level, via a Bayesian hierarchical modelling approach. Results: We present model validation results and anonymised country-level case studies to highlight the impact of including EMU data with uncertainty in FPET when estimating mCPR. The validation results illustrate improved predictive performance with the inclusion of EMUs compared to using survey data only. Case studies provide additional insights into how including EMU data affects mCPR estimates in different country contexts. Together, the validation results and case studies demonstrate that EMUs can help countries more effectively monitor progress toward their family planning goals. 1 Introduction Family planning supports the fundamental right of individuals to choose the number and timing of their children. Access to family planning greatly enhances health outcomes for women and children and helps reduce poverty ( Cleland et al. (2006) ). To effectively track progress toward country-level family planning goals, it is essential for countries to monitor current trends in family planning indicators, such as contraceptive prevalence and unmet need for contraception, and forecast future trends ( Stover and Sonneveldt (2017) ). This empowers countries to make data-driven, informed decisions towards achieving family planning goals. The Family Planning Estimation Tool (FPET) is a Bayesian statistical model that is used to produce country-level estimates and forecasts of family planning indicators ( Alkema et al. (2013) ; Cahill et al. (2018) ; Kantorová et al. (2020) ; Alkema et al. (2024c) ). One of the key family planning indicators used to monitor progress is the modern contraceptive prevalence rate (mCPR), defined as the proportion of women reporting that themselves or their partner currently uses at least one modern contraceptive method. While FPET traditionally relies on survey-based observations of family planning indicators, the intermittent nature of large-scale population-health surveys, conducted every 3-5 years on average, introduces data gaps. This intermittency poses a challenge to achieving data-driven model forecasts and estimates. To address this, we draw on routine health facility data, specifically family planning service statistics, which serve as a supplementary data source generated as a by-product of family planning service delivery. Family planning service statistics are used to derive Estimated Modern Use (EMU), a family planning indicator that provides insight into modern contraceptive use ( Track20 (2020) ; Track20 (2023) ). Despite potential biases in EMU data, studies have demonstrated their utility in FPET to inform mCPR estimates in the absence of recent survey data ( Magnani et al. (2018) ; Cahill et al. (2021) ). To account for such biases, rates of change in EMU data, which are assumed to be unbiased with respect to rates of change in mCPR, can be used to inform mCPR estimates and forecasts where survey data are absent. However, while annual rates of change in EMU estimates have been found to be predictive of true rates of change in mCPR, it has been challenging to capture and quantify the varying sources of uncertainty at the country level associated with this indicator. We present a new approach to using EMUs to inform mCPR estimates in FPET, accounting for both uncertainty associated with the EMU derivation process and the unexplained errors in country-specific EMU data series. Effectively quantifying this uncertainty ultimately improves the use of EMU data in FPET and results in improved accuracy and reliability of the EMU-informed estimates, enabling better tracking of mCPR trends. There is growing interest in the use of non-standard data sources, such as service statistics, to provide frequent and up-to-date insights into population-level health indicators ( Hung et al. (2020) ; Sawadogo-Lewis et al. (2021) ). This is especially useful when informing decision making, particularly where ’gold standard’ data such as household surveys are collected intermittently, not keeping up with the timelines of initiatives, therefore increasing the risk of relying on outdated data. To address this issue, researchers have explored the value of using various forms of non-standard data. For instance, researchers established a spatio-temporal model to combine survey data with routine health data to estimate malaria risk in Rwanda ( Semakula et al. (2023) ). Research into the coverage of maternal and child health services using routine health data further illustrates the use of readily available data to inform indicators ( Maïga et al. (2021) ; Agiraembabazi et al. (2021) ). Additionally, routine health data can offer insights into healthcare performance during crises ( Turcotte-Tremblay et al. (2023) ). These methods can allow for more responsive decision-making, reducing the risk of relying on outdated information. Beyond the use of routine health data, mobile phone data has been used to predict the spatial spread of cholera ( Bengtsson et al. (2015) ), and social media data has been used to track migration patterns ( Alexander et al. (2022) ). The work presented in this paper, improving the use of family planning service statistics to refine estimates of a key family planning indicator such as mCPR, further advances progress in this field. The remainder of this paper is structured as follows: we begin with background, including an overview of FPET. This is followed by a section on service statistics. Next, we present our exploratory analysis and methodology. We then outline the model validation results, discuss the impact of EMU inclusion, and showcase country-level case studies to demonstrate the impact of EMU inclusion across various scenarios. We conclude with a final discussion. 2 Background 2.1 Estimating mCPR using the Family Planning Estimation Tool FPET produces estimates and short-term forecasts of the modern contraceptive prevalence rate (mCPR) for women of reproductive age, by marital status ( Alkema et al. (2013) ; Cahill et al. (2018) ; Kantorová et al. (2020) ; Alkema et al. (2024c) ). mCPR is defined as the proportion of women who are users

Read the complete text