Do cell towers help increase vaccine uptake? Evidence from Côte d’Ivoire

1 Introduction

Vaccination is an effective public health intervention to prevent disease and death. In Africa, around 20% of children fail to receive all basic vaccines, resulting in more than 30 million cases of vaccine-preventable diseases in children under five annually [1]. One of the principal obstacles to vaccination is the difficulty in providing information to parents, especially in rural areas with limited access to education [2]. According to Obi-Jeff et al. [3], the most common reason for non-vaccination is lack of knowledge about where, when, and why to vaccinate.

The rapid increase in mobile technology usage in recent decades has helped overcome obstacles to information access [4,5,6]. Mobile technology reduces information sharing time, facilitates knowledge transfer to less educated populations [7,8], and enhances the efficiency and use of healthcare services [8]. Moreover, it creates educational and economic opportunities and enables access to credit markets, which can lead to better health outcomes [9].

Despite the growing interest in the impact of mobile technology on health outcomes, most research has centered on SMS reminders and mobile phone ownership [10,11,12,13,14,15,16,17,18]. Nonetheless, since phone sharing is common, ownership alone may not accurately reflect access to and usage of mobile communication technology [19]. A systematic review and meta-analysis by Yunusa et al. [20] indicates that the evidence supporting the positive effect of mobile phone interventions on health outcomes is still limited. In addition, some studies linking mobile phone usage with socio-economic outcomes are susceptible to bias because of the existence of unobserved confounders [21].

The current study integrates geocoded cell tower information with childhood vaccination data and finds that closer proximity to cell towers is associated with higher vaccine uptake. To mitigate endogeneity, we build on previous studies and use lightning intensity as an instrumental variable (IV) [21]. The IV estimate aligns with the Ordinary Least Squares (OLS) coefficient, suggesting that proximity to cell phone towers enhances vaccine uptake.

This research provides multiple contributions to the existing body of literature. First, it adds to the existing research on the socio-economic impact of communication infrastructures by exploring their effect on vaccination coverage. Previous studies show that mobile network coverage fosters growth in the retail sector, increases employment, and decreases the prevalence of gender-biased social norms and early marriage [22,23,24]. Some studies have further documented the negative effects of communication infrastructures on property values, suggesting that an increase in distance to cell towers raises property prices [25,26,27]. Other researchers suggest that communication infrastructures facilitate the dissemination of information about ongoing conflicts to remote areas, thereby influencing the diffusion of conflicts [28,29].

Second, we add new insights to the literature regarding the impact of communication infrastructures on health outcomes. Although there are many compelling pathways (see Section 2.1) through which access to cell towers could help increase vaccination coverage, there is a lack of studies analyzing this relationship. An important body of literature has focused on the adverse health effects of communication infrastructures, suggesting that proximity to cell towers increases the likelihood of serious health problems such as cancer and muscular pain [30,31,32,33,34]. Ali et al. investigated the effect of proximity to cell towers on health outcomes and found that people living closer to the cell tower were more likely to have health issues [30]. Evidence from a literature review suggests that proximity to cell towers increases the risk of cancer, changes in biochemical parameters, and radiofrequency-related diseases [31]. Hamza et al. documented that individuals living close to cell towers are more likely to experience muscle pain, fatigue, and headache [32]. The existing literature on positive health effects remains limited, primarily concentrating on HIV testing and infant mortality [21,35].

The rest of this article is organized as follows. Section 2 provides an overview on how cell towers could affect vaccine uptake, the distribution of cell towers, and vaccine uptake in Côte d’Ivoire. Section 3 describes the data and methodology employed in this study. Section 4 presents the OLS results, and Section 5 gives the IV results. Section 6 discusses the results, and Section 7 concludes the study.

2 Background

This section provides an overview of the potential mechanisms through which cell towers could help improve vaccine uptake, cell tower distribution, and how Ivorian authorities have leveraged advances in mobile technology to enhance childhood vaccination coverage.

2.3 Network coverage and vaccination in Côte d’Ivoire

Until the early 2000s, only a few cities outside the economic capital had access to mobile telephone services. However, in the 2010s, the country experienced a significant expansion in network coverage, resulting in a sharp increase in mobile phone ownership. Approximately 81% of households owned at least one mobile phone [45]. As for the Internet, its penetration rate fluctuates around 38% since 2015 [46]. At a median speed of 9.37 Mbps, the average cost for 1 GB of mobile data is about 2.58 USD [47]. Figure 1 illustrates the distribution of cell towers across the country in 2015. It shows some disparities in coverage across the country during our study year.

Figure 1

Distribution of cell towers in Côte d’Ivoire in 2015. Source: Author.

Over the past decades, Ivorian health authorities have leveraged advancements in mobile technology to enhance childhood vaccination rates. In 2011, the country launched its first vaccination reminder service, which involved sending voice and text messages to remind parents of vaccination dates and emphasize their importance. Mobile technology has also been utilized to facilitate communication during vaccination campaigns. For instance, the transparency of the measles vaccination campaign was enhanced through a mobile communication platform that allowed volunteers and families to report issues and obtain information [48]. Another notable example is the use of M vaccine, a mobile application that registers pregnant women and mothers of children under 2 years old, creates digital vaccination records and sends automatic voice reminders in French and local languages to ensure timely vaccinations. It also facilitates transitions from paper-based records to digital systems, allowing for more accurate and efficient tracking of vaccination data [49].

3 Data and methods

4 Results

Table 2 presents our OLS estimates from baseline specifications that include all the vaccines presented in Table 1. All estimates represent the absolute change in our outcome variable. Column 1 shows the estimated effect of mobile phone coverage on vaccine uptake, excluding control variables and fixed effects. The estimated coefficient is negative and statistically significant, suggesting that one unit decrease in the distance to towers increases the probability of a child being vaccinated by about 0.11%. In addition to the control variables described in Table 1, columns 2 and 3 incorporate spatial fixed effects, with column 3 also including vaccine fixed effects. The resulting estimates are similar, indicating that reducing the distance to cell towers by one unit increases the likelihood of a child being vaccinated by approximately 0.14%. This corresponds to approximately two additional children vaccinated per 1,000 inhabitants as the distance decreases by one unit. The results still suggest that proximity to cell towers increases the probability of being vaccinated when we analyze each vaccine separately (Table 3). Nonetheless, the estimated coefficient is the highest for the BCG vaccine, which is associated with the disease having the highest incidence.

We further analyzed the heterogeneities in the association between communication infrastructure access and vaccine uptake. We began by examining the differences between rural and urban residences, with the results presented in Column 1 of Table 4. Both the baseline effect and the coefficient for the interaction between distance to cell towers and rural residence are negative and significant. This indicates that reducing the distance to cell towers significantly increases the likelihood of a child being vaccinated, with a more pronounced effect in rural areas. We also explore differences between low-, middle-, and high-income groups, with high income being the baseline group. Compared with the high-income group, the probability of being vaccinated in the low-income group increases more significantly as the distance from cell towers decreases. However, the result does not indicate any difference between middle- and high-income groups.

5 IV

As previously mentioned, the placement of cell towers is not random; it is influenced by various factors, some of which may correlate with unobserved variables affecting vaccine uptake. Therefore, our OLS estimates may be biased. To mitigate endogeneity issue, we use lightning strike intensity as an IV. The results of the first and second stages are displayed in Panels a and b of Table 5, respectively. The estimates from the first step indicate that higher lightning intensity is associated with increased distance to cell towers. Moreover, the F-statistic is above the threshold, suggesting that our instrument is not weak. In the second step, the IV estimates indicate that reducing the distance to cell towers increases vaccine uptake. Although the coefficient sign is consistent with the OLS estimate, when we exclude the control variables and fixed effects (column 1 of Table 5) the IV estimate is larger in absolute value than the OLS (column 1 of Table 2). However, after including the control variables and fixed effects, the IV estimate (column 2 of Table 5) is similar to the OLS estimates (columns 1 and 2 of Table 2). This may suggest that the inclusion of controls has helped reduce the bias related to potential indirect pathways through which our instrument could affect the outcome variable.

6 Discussion

In this study, we examined the association between access to cell towers and vaccine uptake. We primarily used the distance between household clusters and the closest cell tower to approximate mobile technology access. The results suggest that a decrease in the distance to cell tower increases the probability of a child being vaccinated. The results remained consistent when we used the number of cell towers per 10,000 inhabitants as an alternative proxy for mobile connectivity. These findings align with those of other studies exploring the relationship between communication facilities and health outcomes. According to Iacoella and Tirivayi, a unit decrease in distance from the closest cell tower is associated with 0.5% increase in the probability that women test for HIV [35]. Flückiger and Ludwig found that access to mobile phone coverage reduces childhood mortality risk by 0.9% [21]. The coefficient size in our study is smaller than that found in previous studies. The difference may arise from the fact that these studies employ different outcome variables and focus on multiple countries rather than a specific country in sub-Saharan Africa.

Other findings of interest relate to the heterogeneity analyses. We found that reducing the distance to cell towers increases the likelihood of a child being vaccinated, with a more pronounced effect in rural areas. Moreover, for the low-income group, the probability of being vaccinated increases more significantly as the distance from cell towers decreases than for the high-income group. However, the result does not show any difference between middle and high-income groups. These findings are in line with previous studies, which suggest that the effect of communication infrastructure is greater for people and areas with limited access to information and health services [21,23].

Our study has some limitations that need to be highlighted. First, although our findings suggest a relationship between access to cell towers and vaccine uptake, we are unable to empirically investigate the potential mechanisms due to data constraints. Instead, we discussed about the mechanisms building on previous studies. Second, household location details were unavailable; therefore, we calculated the distance from cluster centroids to the closest cell tower. However, our findings remain consistent when we used the number of cell towers per 10,000 inhabitants. Finally, the cell tower database did not provide construction date of cell towers but rather the date each tower was registered. Hence, there could be a lag between the construction and registration dates. We checked the robustness of our findings by using a sample of cell towers that were registered in the database up to one year after the household survey, and the results aligned with our main findings.

7 Conclusions

This study reveals that better access to mobile technology is associated with increased vaccine uptake, particularly among individuals with limited resources. In light of these findings, we recommend that policymakers collaborate with mobile network operators to deploy cell towers in areas that still lack or have poor connectivity. However, given that cell towers can have adverse health effects, network operators should optimize the distance between homes and cell towers so that they are beneficial without causing harm.