The research team, led by Enoch Kwame Tham-Agyekum of Kwame Nkrumah University of Science and Technology and the University of Pretoria, set out to answer a deceptively simple question: who actually gets access to climate information broadcast over community radio, and who is left out? Rather than treating access as a yes-or-no matter of owning a radio, the researchers conceptualized it as a four-dimensional construct spanning physical, socio-cultural, economic, and institutional or policy-related accessibility. This framework allowed them to disentangle the many ways a farmer can be cut off from the airwaves, from missing batteries and bad roads to gender norms that keep women away from listening groups.

The scale of the study is notable. Working from an estimated population of roughly 870,000 smallholder farmers across the Northern, Upper East, Upper West, Savannah, and North East regions, the team surveyed 1,200 farmers, more than three times the minimum sample size required for statistical reliability at a 95 percent confidence level. Respondents were selected through multi-stage sampling, with regions, radio stations, and communities chosen purposively for their climate vulnerability and active broadcasting, and individual farmers drawn randomly in proportion to each region’s farming population. The survey data were then complemented by interviews with 25 key informants, including agricultural extension agents, religious and traditional leaders, community leaders, and NGO representatives.

What the researchers found first was how deeply community radio is woven into daily agricultural life. More than 58 percent of farmers reported listening to climate-related broadcasts every day, and another 24.5 percent tuned in weekly. Only about 6 percent never listened at all. Most farmers used their own radio sets, at 44.25 percent, or mobile phones, at 37.83 percent, while roughly one in five relied on neighbors’ radios or communal listening groups. The interviews revealed that these shared devices matter enormously in remote areas, where farmers gather to hear a forecast and immediately discuss what it means for planting dates.

The content itself proved far richer than simple weather reports. Key informants described broadcasts covering seasonal forecasts, rainfall warnings, planting-date guidance, crop diversification, drought-resistant seed varieties, soil management, post-harvest storage, livestock care during dry spells, and even market prices during drought. The authors characterize community radio as an integrated climate information system, one that combines early warning, advisory services, indigenous knowledge such as traditional signs of changing weather, and market intelligence across the entire agricultural value chain. In a region where formal extension services are often understaffed and unable to reach remote communities, this makes radio a lifeline for climate-smart decision-making.

But the study’s most striking findings concern the patterns of exclusion hiding inside those impressive listenership numbers. Using a Multivariate Probit model, which allowed the researchers to estimate the four access dimensions simultaneously while accounting for correlations among them, the team identified clear socio-economic fault lines. Gender emerged as a powerful determinant: being female was associated with significantly lower physical, economic, and institutional accessibility, with the physical dimension significant at the p < 0.01 level and the economic and institutional dimensions at p < 0.05. The authors attribute this to mobility constraints, domestic responsibilities, and lower device ownership among women, citing comparative evidence from Benin where 67 percent of women rice processors owned a radio set compared with 87 percent of men.

Household size told a similar story. Farmers from larger households, which in this sample averaged about nine members, experienced significantly lower physical, economic, and institutional accessibility, all at p < 0.01. The explanation is partly economic: a single radio set shared among many family members limits individual access, and the time and resource pressures of feeding a large household squeeze out the costs of batteries, charging, and participation in information networks. Extension access, by contrast, was the strongest positive predictor, significantly boosting physical, economic, and institutional accessibility, confirming that farmers already connected to agricultural extension agents are far more likely to benefit from radio-linked institutional programming. Higher monthly income also improved accessibility across three dimensions.

One result defied expectations: access to formal credit was negatively associated with all four dimensions of accessibility at p < 0.01. The researchers offer several plausible explanations rather than a definitive one. Farmers with credit may rely on alternative, more personalized information channels such as extension personnel, input suppliers, or mobile-based services, making community radio relatively less important to them. Credit obligations may also reshape how farmers allocate their time, pulling them away from community-based information channels. Whatever the mechanism, the finding underscores that financial resources alone do not guarantee engagement with agricultural knowledge systems, particularly where coordination between financial institutions and extension services remains weak.

The perception indices added texture to the statistical picture. Physical accessibility scored highest in farmers’ estimations at 3.69 on a five-point scale, with station locations and broadcast schedules well aligned to farming routines, though rainy-season roads and patchy signals in interior communities posed real problems. Institutional and policy accessibility followed at 3.75, buoyed by strong collaboration between extension agents and radio stations, which scored 4.31, but dragged down by weak NGO support and inconsistent alignment with government agricultural policy. Socio-cultural accessibility scored 3.60, with farmers praising local-language broadcasts that make complex forecasts understandable, yet noting that chiefs and elders, the community’s most trusted voices, are too rarely involved on air. Economic accessibility was the weakest dimension at 3.50, with informants reporting that even small costs become prohibitive during the lean season and that some households lack reliable radios or batteries altogether.

The study’s recommendations are as concrete as its diagnosis. The authors call for gender-responsive listening groups supported with solar-powered radios for women, broadcast schedules aligned with women’s domestic and farming responsibilities, and formal integration of traditional leaders into climate programming to build trust and message acceptance. They propose targeted subsidies or cost-sharing schemes to boost radio ownership among credit-constrained and large-household farmers, multi-stakeholder content committees to improve the technical depth and policy alignment of broadcasts, and integration of radio with mobile-phone services such as SMS alerts and call-in platforms. Each proposal comes with honest caveats about funding sustainability, network coverage, and digital literacy gaps. The broader lesson reaches beyond Ghana: as climate change intensifies pressure on rain-fed agriculture across the Sahel and beyond, the channels that deliver life-saving forecasts are only as effective as their weakest link, and ensuring that women, the poor, and the marginalized can actually hear the message may matter as much as the accuracy of the forecast itself.

Subject of Research: Inclusion and exclusion in climate information access through community radio among smallholder farmers in Northern Ghana

Article Title: Patterns of inclusion and exclusion in climate information access through community radio among smallholder farmers in Northern Ghana

Article References: Tham-Agyekum, E. K., Nimoh, F., Assani, A. S., Maramura, T. C., Cheboi, J. J., Wainaina, R. W., & Bakang, J.-E. A. (2026). Patterns of inclusion and exclusion in climate information access through community radio among smallholder farmers in Northern Ghana. Discover Agriculture, 4(1), Article 310. https://doi.org/10.1007/s44279-026-00792-0

Image Credits: AI Generated

DOI: 10.1007/s44279-026-00792-0

Keywords: community radio, climate information, smallholder farmers, Northern Ghana, climate adaptation, gender inequality, agricultural extension, information accessibility, rain-fed agriculture, climate resilience, mobile phones, rural communication

The review team, led by Maggie Wang of Cornell’s Department of Nutritional Sciences together with senior author Ka Kahe of Columbia University Irving Medical Center, approached the problem with the formal machinery of a modern systematic review. Comprehensive searches of PubMed, Web of Science and Google Scholar were conducted, and the Covidence platform was used for title and abstract screening against predetermined inclusion criteria. Out of the published literature, twenty studies met the bar for inclusion. The authors applied established risk-of-bias tools, including versions of the Cochrane RoB 2 and ROBINS-I instruments documented in their supplementary materials, and followed PRISMA 2020 reporting guidelines. The focus was deliberately narrow: human studies of maternal mobile phone use during pregnancy, with exposure characterized by how often and for how long women used their phones, and outcomes ranging from miscarriage and gestational length to fetal growth, congenital anomalies and newborn measurements.

Why should anyone expect phones to affect pregnancy at all? The physics and biology behind the concern are worth spelling out. Mobile phones communicate via non-ionizing radiofrequency electromagnetic fields, or RF-EMF, which lack the energy to break chemical bonds or damage DNA directly in the way that ionizing radiation can. However, two broad categories of biological mechanism have been proposed. The first is thermal: RF-EMF energy absorbed by tissue raises its temperature slightly, and the standard metric for this is the specific absorption rate, which regulators such as the FCC and ICNIRP use to set exposure limits. The second, more contentious category is non-thermal: laboratory work has reported that low-intensity RF exposure can elevate reactive oxygen species, alter antioxidant enzyme activity, and change protein expression in placental tissue, even without measurable heating. Pregnancy is precisely the window when such subtle perturbations could matter most, because fetal development involves rapid cell division, organogenesis and cardiovascular adaptation, all of which are known to be sensitive to oxidative stress and other environmental influences.

The oxidative stress pathway is particularly relevant to the two outcomes the review flagged. Fetal developmental physiology studies have shown that prenatal hypoxia and oxidative stress can program long-term cardiovascular dysfunction, while work on human fetal membranes has linked oxidative damage to premature senescence and preterm premature rupture of membranes. If RF-EMF exposure indeed tips redox balance in placental or fetal tissue even modestly, the cardiovascular system and the timing of birth are plausible downstream targets. This biological plausibility does not prove causation, but it explains why the review’s authors argue that system-specific vulnerabilities deserve dedicated research rather than broad, generic reassurance.

What does the human evidence actually show? For fetal and neonatal cardiac function, several included studies observed measurable changes. A Turkish study published in 2004 in the European Journal of Obstetrics & Gynecology and Reproductive Biology found that a phone call made during pregnancy altered fetal heart rate patterns recorded by cardiotocography. A larger Egyptian investigation reported in Saudi Medical Journal in 2008 assessed fetal and neonatal responses following maternal mobile phone exposure and documented differences in cardiac parameters between exposed and less-exposed groups, while a BJOG conference report examined antenatal cardiotocography at term in relation to phone radiation. These findings were not uniformly replicated across the literature, but the reviewers judged them to be the most consistent signals in the entire dataset. Because heart rate variability and fetal cardiac reactivity are sensitive indicators of autonomic function, even transient changes raise questions that conventional retrospective study designs cannot easily answer.

The second consistent finding concerned gestational duration. Some studies in the review reported an association between heavier maternal mobile phone use and an increased risk of preterm birth, the leading cause of neonatal morbidity worldwide. This result stood in contrast to outcomes such as birth weight, fetal growth measurements and congenital anomalies, where the twenty studies produced mixed and frequently null findings. A major multi-cohort analysis of four birth cohorts published in the American Journal of Epidemiology in 2019 examined maternal cell phone use during pregnancy against both pregnancy duration and fetal growth and found the picture far from alarming, illustrating the broader pattern: for many outcomes, including miscarriage and newborn anthropometrics, individual studies disagreed, and the review declined to declare a clear risk.

That inconsistency is itself the central scientific problem. The authors identify a long list of methodological limitations that have plagued this field for two decades. Exposure assessment in nearly every study relies on self-reported call frequency or duration, which correlates poorly with actual RF-EMF dose; modern phones in data mode can produce peak exposures that differ dramatically from voice mode, and real-world exposure varies enormously with network conditions, device positioning and standby behavior. No included study measured personal RF-EMF exposure with dosimeters. Confounding is another hazard: heavy phone users differ from light users in socioeconomic status, sedentary behavior, sleep, stress and other factors that independently influence pregnancy outcomes. The review also notes that phone technology has shifted from 2G through 4G to 5G, changing exposure patterns faster than epidemiology can track them, and that experimental studies often use simulated exposures that do not faithfully reproduce real emissions.

The reviewers are explicit about what should happen next. They call for standardized, quantitative exposure assessment—ideally using validated dosimetry or modeling rather than questionnaires—and for large, well-designed prospective cohort studies with repeated exposure measurements across pregnancy. They also argue for outcome-specific research on fetal cardiac function and preterm birth, the two domains where the existing evidence clusters most coherently. Such work would allow dose-response modeling, which the current sparse and heterogeneous literature cannot support. In parallel, they suggest that experimental studies should use realistic exposure scenarios and focus on mechanistic endpoints such as oxidative stress markers in placental tissue, where preliminary human studies of cord blood and placenta have already reported shifts in oxidant and antioxidant levels.

For pregnant women and clinicians, the practical takeaway is one of informed nuance rather than alarm. The review does not conclude that mobile phones are safe in pregnancy, nor that they are dangerous; it concludes that the human evidence is inconclusive, with two specific domains—fetal cardiac function and gestational duration—where the data justify focused concern and further study. Regulatory limits based on thermal effects remain in force, but the authors argue, consistent with a small group of exposure scientists, that the health assumptions underpinning current FCC and ICNIRP limits deserve renewed scrutiny for susceptible populations such as pregnant women. Until stronger data arrive, simple precautionary measures—holding the phone slightly away from the abdomen, using speaker mode or wired headsets for long calls, and limiting prolonged data-mode exposure near the body—remain low-cost options that many practitioners may reasonably suggest, without a basis for prescriptive rules. What this review ultimately delivers is a roadmap: it synthesizes the existing human evidence honestly, exposes the gaps that have kept the debate alive for twenty years, and sets out the specific, testable questions that must be answered before science can tell expectant mothers, with confidence, what their phones are and are not doing to their pregnancies.

Subject of Research: Associations between maternal mobile phone use and pregnancy and fetal outcomes

Article Title: A systematic review of mobile phone use and pregnancy and fetal outcomes

Article References: A systematic review of mobile phone use and pregnancy and fetal outcomes. (n.d.). https://doi.org/10.1038/s41370-026-00958-4

Image Credits: AI Generated

DOI: 10.1038/s41370-026-00958-4

Keywords: mobile phones, pregnancy outcomes, fetal health, RF-EMF exposure, preterm birth, fetal cardiac function, systematic review, non-ionizing radiation, maternal health, oxidative stress, epidemiology, radiofrequency electromagnetic fields