Evaluating food security and nutritional pathways of rural farm families: Empirical evidence from northeast India.

The study aimed to ascertain a relationship between agricultural status, socioeconomic factors, and nutrition of farm families. The study was conducted in selected villages in the West Garo Hills district of Meghalaya, using Stratified Random Sampling (St. RS). Using pretested interview schedules, we collected primary data from respondents in 2020 and 2021, focusing on socioeconomic variables, body mass index, and income from agriculture and related sectors. The data was analysed using correlation analyses and separate combined regression estimates for each year and month were obtained. Results from the study indicate that agricultural income significantly influenced nutritional status (p < 0.05) and household income growth was also found significant. The region's agricultural production of cereals, pulses, and vegetables was insufficient, as was the production of meat and meat products, milk, and milk products. Hence, expenditure towards purchasing the above food groups from the market was found to be significant (p < 0.05). Therefore, the markets near the mainland especially in the hilly region play a crucial role in the nutritional pathway of rural farm families.

Fitting a progressive three-state colorectal cancer model to interval-censored surveillance data under outcome-dependent sampling using a weighted likelihood approach.

To optimize colorectal cancer (CRC) surveillance, accurate information on the risk of developing CRC from premalignant lesions is essential. However, directly observing this risk is challenging since precursor lesions, i.e., advanced adenomas (AAs), are removed upon detection. Statistical methods for multistate models can estimate risks, but estimation is challenging due to low CRC incidence. We propose an outcome-dependent sampling (ODS) design for this problem in which we oversample CRCs. More specifically, we propose a three-state model for jointly estimating the time distributions from baseline colonoscopy to AA and from AA onset to CRC accounting for the ODS design using a weighted likelihood approach. We applied the methodology to a sample from a Norwegian adenoma cohort (1993-2007), comprising 1, 495 individuals (median follow-up 6.8 years [IQR: 1.1 - 12.8 years]) of whom 648 did and 847 did not develop CRC. We observed a 5-year AA risk of 13% and 34% for individuals having non-advanced adenoma (NAA) and AA removed at baseline colonoscopy, respectively. Upon AA development, the subsequent risk to develop CRC in 5 years was 17% and age-dependent. These estimates provide a basis for optimizing surveillance intensity and determining the optimal trade-off between CRC prevention, costs, and use of colonoscopy resources.

Global biomass maps can increase the precision of (sub)national aboveground biomass estimates: A comparison across tropical countries.

Countries within the tropics face ongoing challenges in completing or updating their national forest inventories (NFIs), critical for estimating aboveground biomass (AGB) and for forest-related greenhouse gas (GHG) accounting. While previous studies have explored the integration of map information with local reference data to fill in data gaps, limited attention has been given to the specific challenges presented by the clustered plot designs frequently employed by NFIs when combined with remote sensing-based biomass map units. This research addresses these complexities by conducting four country case-studies, encompassing a variety of NFI characteristics within a range of AGB densities. Examining four country case-studies (Peru, Guyana, Tanzania, Mozambique), we assess the potential of European Space Agency's Climate Change Initiative (CCI) global biomass maps to increase precision in (sub)national AGB estimates. We compare a baseline approach using NFI field-based data with a model-assisted scenario incorporating a locally calibrated CCI biomass map as auxiliary information. The original CCI biomass maps systematically underestimate AGB in three of the four countries at both the country and stratum level, with particularly weak agreement at finer map resolution. However, after calibration with country-specific NFI data, stratum and country-level AGB estimates from the model-assisted scenario align well with those obtained solely from field-based data and official country reports. Introducing maps as a source of auxiliary information fairly increased the precision of stratum and country-wise AGB estimates, offering greater confidence in estimating AGB for GHG reporting purposes. Considering the challenges tropical countries face with implementing their NFIs, it is sensible to explore the potential benefits of biomass maps for climate change reporting mechanisms across biomes. While country-specific NFI design assumptions guided our model-assisted inference strategies, this study also uncovers transferable insights from the application of global biomass maps with NFI data, providing valuable lessons for climate research and policy communities.

Efficient risk-based collection of biospecimens in cohort studies: Designing a prospective study of diagnostic performance for multicancer detection tests.

In cohort studies, it can be infeasible to collect specimens on an entire cohort. For example, to estimate sensitivity of multiple Multi-Cancer Detection (MCD) assays, we desire an extra 80mL of cell-free DNA (cfDNA) blood, but this much extra blood is too expensive for us to collect on everyone. We propose a novel epidemiologic study design that efficiently oversamples those at highest baseline disease risk from whom to collect specimens, to increase the number of future cases with cfDNA blood collection. The variance reduction ratio from our risk-based subsample versus a simple random (sub)sample (SRS) depends primarily on the ratio of risk model sensitivity to the fraction of the cohort selected for specimen collection subject to constraining the risk model specificity. In a simulation where we chose 34% of Prostate, Lung, Colorectal, and Ovarian Screening Trial cohort at highest risk of lung cancer for cfDNA blood collection, we could enrich the number of lung cancers 2.42-fold and the standard deviation of lung-cancer MCD sensitivity was 31-33% reduced versus SRS. Risk-based collection of specimens on a subsample of the cohort could be a feasible and efficient approach to collecting extra specimens for molecular epidemiology.

Six Steps towards a Spatial Design for Large-Scale Pollinator Surveillance Monitoring.

Despite the importance of pollinators to ecosystem functioning and human food production, comprehensive pollinator monitoring data are still lacking across most regions of the world. Policy-makers have recently prioritised the development of large-scale monitoring programmes for pollinators to better understand how populations respond to land use, environmental change and restoration measures in the long term. Designing such a monitoring programme is challenging, partly because it requires both ecological knowledge and advanced knowledge in sampling design. This study aims to develop a conceptual framework to facilitate the spatial sampling design of large-scale surveillance monitoring. The system is designed to detect changes in pollinator species abundances and richness, focusing on temperate agroecosystems. The sampling design needs to be scientifically robust to address questions of agri-environmental policy at the scales of interest. To this end, we followed a six-step procedure as follows: (1) defining the spatial sampling units, (2) defining and delimiting the monitoring area, (3) deciding on the general sampling strategy, (4) determining the sample size, (5) specifying the sampling units per sampling interval, and (6) specifying the pollinator survey plots within each sampling unit. As a case study, we apply this framework to the "Wild bee monitoring in agricultural landscapes of Germany" programme. We suggest this six-step procedure as a conceptual guideline for the spatial sampling design of future large-scale pollinator monitoring initiatives.

Deficiencies in monitoring practices of marine protected areas in southern European seas.

Worldwide, states are gazetting new Marine Protected Areas (MPAs) to meet the international commitment of protecting 30% of the seas by 2030. Yet, protection benefits only come into effect when an MPA is implemented with activated regulations and actively managed through continuous monitoring and adaptive management. To assess if actively managed MPAs are the rule or the exception, we used the Mediterranean and Black Seas as a case study, and retrieved information on monitoring activities for 878 designated MPAs in ten European Union (EU) countries. We searched for scientific and grey literature that provides information on the following aspects of MPA assessment and monitoring: ecological (e.g., biomass of commercially exploited fish), social (e.g., perceptions of fishers in an MPA), economic (e.g., revenue of fishers) and governance (e.g., type of governance scheme). We also queried MPA authorities on their past and current monitoring activities using a web-based survey through which we collected 123 responses. Combining the literature review and survey results, we found that approximately 16% of the MPA designations (N = 878) have baseline and/or monitoring studies. Most monitoring programs evaluated MPAs based solely on biological/ecological variables and fewer included social, economic and/or governance variables, failing to capture and assess the social-ecological dimension of marine conservation. To increase the capacity of MPAs to design and implement effective social-ecological monitoring programs, we recommend strategies revolving around three pillars: funding, collaboration, and technology. Following the actionable recommendations presented herein, MPA authorities and EU Member States could improve the low level of MPA monitoring to more effectively reach the 30% protection target delivering benefits for biodiversity conservation.

A three-dimensional sampling design based on the coefficient of variation method for soil environmental damage investigation.

A traditional grid model for soil sampling may suffer from poor efficiency and low accuracy. With a nonferrous metal processing plant as the study area, a three-dimensional kriging interpolation model was built based on this plant's preliminary investigation data for arsenic (As), and a detailed survey sampling programme was proposed. The sampling density at the pollution interval of the surface soil was estimated by the coefficient of variation method, and the sampling depth was determined by the pollution interval of the vertical prediction results. The results showed that the encrypted soil sampling distribution optimisation method obtains greater pointing accuracy with fewer points. The sampling accuracy was 87.62% after optimising the depth of pointing. Moreover, this approach could save 66.13% of the sampling costs and 56.93% of the testing costs compared to a full deployment programme. This study provides a new and cost-effective method for predicting the extent of contamination exceedance at a site and provides valuable information to guide post-remediation strategies for contaminated sites.

Separating Measurement Error and Signal in Environmental Data: Use of Replicates to Address Uncertainty.

Measurement uncertainty has long been a concern in the characterizing and interpreting environmental and toxicological measurements. We compared statistical analysis approaches when there are replicates: a Naïve approach that omits replicates, a Hybrid approach that inappropriately treats replicates as independent samples, and a Measurement Error Model (MEM) approach in a random effects analysis of variance (ANOVA) model that appropriately incorporates replicates. A simulation study assessed the effects of sample size and levels of replication, signal variance, and measurement error on estimates from the three statistical approaches. MEM results were superior overall with confidence intervals for the observed mean narrower on average than those from the Naïve approach, giving improved characterization. The MEM approach also featured an unparalleled advantage in estimating signal and measurement error variance separately, directly addressing measurement uncertainty. These MEM estimates were approximately unbiased on average with more replication and larger sample sizes. Case studies illustrated analyzing normally distributed arsenic and log-normally distributed chromium concentrations in tap water and calculating MEM confidence intervals for the true, latent signal mean and latent signal geometric mean (i.e., with measurement error removed). MEM estimates are valuable for study planning; we used simulation to compare various sample sizes and levels of replication.

In models we trust: preregistration, large samples, and replication may not suffice.

Despite discussions about the replicability of findings in psychological research, two issues have been largely ignored: selection mechanisms and model assumptions. Both topics address the same fundamental question: Does the chosen statistical analysis tool adequately model the data generation process? In this article, we address both issues and show, in a first step, that in the face of selective samples and contrary to common practice, the validity of inferences, even when based on experimental designs, can be claimed without further justification and adaptation of standard methods only in very specific situations. We then broaden our perspective to discuss consequences of violated assumptions in linear models in the context of psychological research in general and in generalized linear mixed models as used in item response theory. These types of misspecification are oftentimes ignored in the psychological research literature. It is emphasized that the above problems cannot be overcome by strategies such as preregistration, large samples, replications, or a ban on testing null hypotheses. To avoid biased conclusions, we briefly discuss tools such as model diagnostics, statistical methods to compensate for selectivity and semi- or non-parametric estimation. At a more fundamental level, however, a twofold strategy seems indispensable: (1) iterative, cumulative theory development based on statistical methods with theoretically justified assumptions, and (2) empirical research on variables that affect (self-) selection into the observed part of the sample and the use of this information to compensate for selectivity.

Longitudinal course and outcome of social jetlag in adolescents: A 1-year follow-up study of the adolescent sleep health epidemiological cohorts.

The discrepancy in sleep timing between weekdays and weekends - social jetlag (SJL) - is known to negatively affect student quality of life (QOL). However, the association between social jetlag and physical/mental QOL among adolescents and the precise effect of social jetlag on depressive symptoms and daytime sleepiness remains unknown. This study investigated the longitudinal course, risk factors, and effects of social jetlag, a circadian misalignment, in a school-based cohort. The participants were 427 students (13.3 ± 0.6 years, 45.2% girls) from five junior high schools. We performed a baseline survey in 2019 and a 1-year follow-up survey in 2020. Depressive symptoms, QOL, and daytime sleepiness were assessed using the Birleson Depression Self-Rating Scale for Children, Paediatric Quality of Life Inventory, and Paediatric Daytime Sleepiness Scale. In the baseline survey, 49.6% of the students reported SJL ≥1 h, and 17.1% reported SJL ≥2 h. Among them, 37.2% and 6.8% reported persistent SJL at follow-up, respectively. New incidences of SJL ≥1 h were associated with older age, non-attainment of menarche or voice changes, and longer duration of smartphone use, whereas its persistence was associated with a later chronotype. Persistence of SJL ≥1 h and ≥2 h predicted depressive symptoms and daytime sleepiness at follow-up, whereas new incidences of SJL ≥2 h predicted lower QOL. In conclusion, social jetlag has a persistent course, and daytime functioning can deteriorate as social jetlag becomes chronic. Our findings suggest the need for intensive interventions for social jetlag among adolescents.

Sampling effect in predicting the evolutionary response of populations to climate change.

Genomic data and machine learning approaches have gained interest due to their potential to identify adaptive genetic variation across populations and to assess species vulnerability to climate change. By identifying gene-environment associations for putatively adaptive loci, these approaches project changes to adaptive genetic composition as a function of future climate change (genetic offsets), which are interpreted as measuring the future maladaptation of populations due to climate change. In principle, higher genetic offsets relate to increased population vulnerability and therefore can be used to set priorities for conservation and management. However, it is not clear how sensitive these metrics are to the intensity of population and individual sampling. Here, we use five genomic datasets with varying numbers of SNPs (NSNPs = 7006-1,398,773), sampled populations (Npop = 23-47) and individuals (Nind = 185-595) to evaluate the estimation sensitivity of genetic offsets to varying degrees of sampling intensity. We found that genetic offsets are sensitive to the number of populations being sampled, especially with less than 10 populations and when genetic structure is high. We also found that the number of individuals sampled per population had small effects on the estimation of genetic offsets, with more robust results when five or more individuals are sampled. Finally, uncertainty associated with the use of different future climate scenarios slightly increased estimation uncertainty in the genetic offsets. Our results suggest that sampling efforts should focus on increasing the number of populations, rather than the number of individuals per populations, and that multiple future climate scenarios should be evaluated to ascertain estimation sensitivity.

Optimal two-stage spatial sampling design for estimating critical parameters of SARS-CoV-2 epidemic: Efficiency versus feasibility.

The COVID-19 pandemic presents an unprecedented clinical and healthcare challenge for the many medical researchers who are attempting to prevent its worldwide spread. It also presents a challenge for statisticians involved in designing appropriate sampling plans to estimate the crucial parameters of the pandemic. These plans are necessary for monitoring and surveillance of the phenomenon and evaluating health policies. In this respect, we can use spatial information and aggregate data regarding the number of verified infections (either hospitalized or in compulsory quarantine) to improve the standard two-stage sampling design broadly adopted for studying human populations. We present an optimal spatial sampling design based on spatially balanced sampling techniques. We prove its relative performance analytically in comparison to other competing sampling plans, and we also study its properties through a series of Monte Carlo experiments. Considering the optimal theoretical properties of the proposed sampling plan and its feasibility, we discuss suboptimal designs that approximate well optimality and are more readily applicable.

When case reporting becomes untenable: Can sewer networks tell us where COVID-19 transmission occurs?

Monitoring SARS-CoV-2 in wastewater is a valuable approach to track COVID-19 transmission. Designing wastewater surveillance (WWS) with representative sampling sites and quantifiable results requires knowledge of the sewerage system and virus fate and transport. We developed a multi-level WWS system to track COVID-19 in Atlanta using an adaptive nested sampling strategy. From March 2021 to April 2022, 868 wastewater samples were collected from influent lines to wastewater treatment facilities and upstream community manholes. Variations in SARS-CoV-2 concentrations in influent line samples preceded similar variations in numbers of reported COVID-19 cases in the corresponding catchment areas. Community sites under nested sampling represented mutually-exclusive catchment areas. Community sites with high SARS-CoV-2 detection rates in wastewater covered high COVID-19 incidence areas, and adaptive sampling enabled identification and tracing of COVID-19 hotspots. This study demonstrates how a well-designed WWS provides actionable information including early warning of surges in cases and identification of disease hotspots.

Optimal sample type and number vary in small shallow lakes when targeting non-native fish environmental DNA.

Non-native fish have been shown to have deleterious impacts on freshwater ecosystems in New Zealand. Early detection is critical for their effective management. Traditional capture-based techniques may not detect newly introduced fish, especially if they are present in low abundance. Molecular techniques that target environmental DNA (eDNA) have been shown, in many instances, to be more sensitive, cost-effective and require lower sampling effort. However, appropriate sampling strategies are needed to ensure robust and interpretable data are obtained. In this study we used droplet digital PCR assays to investigate the presence of two non-native fish in New Zealand, the European perch (Perca fluviatilis) and rudd (Scardinius erythrophthalmus) in three small lakes. Samples were collected from water and surface sediment at near-shore and mid-lake sites. Probabilistic modelling was used to assess the occupancy of fish eDNA and develop guidance on sampling strategies. Based on the detection probability measures from the present study, at least six sites and five replicates per site are needed to reliably detect fish eDNA in sediment samples, and twelve sites with eight replicates per site for water samples. The results highlight the potential of developing monitoring and surveillance programs adapted to lakes, that include the use of assays targeting eDNA. This study focused on small shallow lakes, and it is likely that these recommendations may vary in larger, deeper, and more geomorphologically complex lakes, and this requires further research.

Comparisons of statistical methods for handling attrition in a follow-up visit with complex survey sampling.

Design-based analysis, which accounts for the design features of the study, is commonly used to conduct data analysis in studies with complex survey sampling, such as the Hispanic Community Health Study/Study of Latinos (HCHS/SOL). In this type of longitudinal study, attrition has often been a problem. Although there have been various statistical approaches proposed to handle attrition, such as inverse probability weighting (IPW), non-response cell weighting (NRCW), multiple imputation (MI), and full information maximum likelihood (FIML) approach, there has not been a systematic assessment of these methods to compare their performance in design-based analyses. In this article, we perform extensive simulation studies and compare the performance of different missing data methods in linear and generalized linear population models, and under different missing data mechanism. We find that the design-based analysis is able to produce valid estimation and statistical inference when the missing data are handled appropriately using IPW, NRCW, MI, or FIML approach under missing-completely-at-random or missing-at-random missing mechanism and when the missingness model is correctly specified or over-specified. We also illustrate the use of these methods using data from HCHS/SOL.

Estimating the population size of a mountain galliform in the context of multi-stakeholder adaptive management.

We designed a participatory monitoring program for the capercaillie population in the French Pyrenees based on lek censuses conducted during the breeding season. This program was implemented by a consortium of stakeholders interested in the conservation of French galliforms. The program, carried out since 2010, relied on a dual frame sampling approach: The first sampled frame was the list of all known leks in the study area. We distinguished two types of known leks: leks known to be active before the onset of the program (with at least one cock detected since 2000) and leks with an indeterminate activity status at the time of the onset of the program. The monitoring program also accounted for the existence of leks that were unknown due mainly to incomplete expert knowledge. We therefore built a complementary area frame by discretizing the study area into a set of 4-km2 grid cells. These cells were then sampled and searched to find unknown leks. When unknown leks were found, cock censuses were organized. An additional field experiment allowed us to estimate the detection probability of unknown leks during these cell searches. We then fitted two hierarchical models: (i) An N-mixture model fitted to the lek census data set allowed us to estimate the mean number of cocks on the three types of leks (known active, known indeterminate, and unknown leks); and (ii) another model fitted to the cell search data set allowed us to estimate the number of unknown leks in the studied mountain range. By multiplying the estimated mean numbers of cocks associated with the three types of leks by the number of leks of each type (an estimated value in the case of unknown leks), we obtained estimates of the total numbers of cocks on all leks at different spatial scales in the study area every 2 years. Our model suggests that the capercaillie cock population was stable from 2010 to 2017 over the whole range but decreased slightly in the foothill area and western part, a decrease that worsened in 2018-2019.

Net overboard: Comparing marine eDNA sampling methodologies at sea to unravel marine biodiversity.

Environmental DNA (eDNA) analyses are powerful for describing marine biodiversity but must be optimized for their effective use in routine monitoring. To maximize eDNA detection probabilities of sparsely distributed populations, water samples are usually concentrated from larger volumes and filtered using fine-pore membranes, often a significant cost-time bottleneck in the workflow. This study aimed to streamline eDNA sampling by investigating plankton net versus bucket sampling, direct versus sequential filtration including self-preserving filters. Biodiversity was assessed using metabarcoding of the small ribosomal subunit (18S rRNA) and mitochondrial cytochrome c oxidase I (COI) genes. Multispecies detection probabilities were estimated for each workflow using a probabilistic occupancy modelling approach. Significant workflow-related differences in biodiversity metrics were reported. Highest amplicon sequence variant (ASV) richness was attained by the bucket sampling combined with self-preserving filters, comprising a large portion of microplankton. Less diversity but more metazoan taxa were captured in the net samples combined with 5 µm pore size filters. Prefiltered 1.2 µm samples yielded few or no unique ASVs. The highest average (~32%) metazoan detection probabilities in the 5 µm pore size net samples confirmed the effectiveness of preconcentration plankton for biodiversity screening. These results contribute to streamlining eDNA sampling protocols for uptake and implementation in marine biodiversity research and surveillance.

Re-thinking the environment in landscape genomics.

Detecting the extrinsic selective pressures shaping genomic variation is critical for a better understanding of adaptation and for forecasting evolutionary responses of natural populations to changing environmental conditions. With increasing availability of geo-referenced environmental data, landscape genomics provides unprecedented insights into how genomic variation and underlying gene functions affect traits potentially under selection. Yet, the robustness of genotype-environment associations used in landscape genomics remains tempered due to various limitations, including the characteristics of environmental data used, sampling designs employed, and statistical frameworks applied. Here, we argue that using complementary or new environmental data sources and well-informed sampling designs may help improve the detection of selective pressures underlying patterns of local adaptation in various organisms and environments.

Multilevel analysis of individual- and community-level determinants of birth certification of children under-5 years in Nigeria: evidence from a household survey.

Promoting birth certification is central to achieving legal identity for all - target 16.9 of the 2030 Sustainable Development Goals. Nigeria is not on track to achieve this goal with its low coverage of birth certification (BC). This study is aimed at identifying patterns of BC and its associated individual- and community-level factors, using pooled cross-sectional data from three rounds (2008, 2013, and 2018) of the nationally representative Nigerian Demographic and Health Survey. A weighted sample of 66,630 children aged 0-4 years was included, and a two-level multilevel logistic model which accommodates the hierarchical nature of the data was employed. Of the total sample, 17.1% [95% CI: 16.3-17.9] were reported to be certified. Zamfara state (2.3, 95% CI: 0.93-3.73) and the Federal Capital Territory (36.24, 95% CI: 31.16-41.31) reported the lowest and the highest BC rates. Children with an SBA [AOR = 1.283, 95% CI: 1.164-1.413] and with at least one vaccination [AOR = 1.494, 95% CI: 1.328-1.681] had higher odds of BC. The AOR for mothers with at least one prenatal visit was 1.468 [95% CI: 1.271-1.695], and those aged 30-34 years at the time of birth [AOR = 1.479, 95% CI: 1.236-1.772] had the highest odds. Further, the odds of BC increased the most for mothers [AOR = 1.559, 95% CI: 1.329-1.829] and fathers [AOR = 1.394, 95% CI: 1.211-1.605] who were tertiary-educated. In addition, children in middle-income [AOR = 1.430, 95% CI: 1.197-1.707] or rich wealth HHs [AOR = 1.776, 95% CI: 1.455-2.169] or those whose families had bank accounts [AOR = 1.315, 95% CI: 1.187-1.456] had higher odds. Living in non-poor and within close proximity to a registration center (RC) act as protective factors for BC, while living in poor communities [AOR = 0.613, 95% CI: 0.486-0.774] and more than 10kms from an RC reduce the odds of BC [AOR = 0.466, 95% CI: 0.377-0.576]. The study identified several protective and risk factors which policymakers can adopt as strategic areas for universal birth certification. National and sub-national programs should integrate non-formal institutions as well as target child and maternal utilization of healthcare services to promote BC in Nigeria.

Perceived multiple stressor effects depend on sample size and stressor gradient length.

Multiple stressors are continuously deteriorating surface waters worldwide, posing many challenges for their conservation and restoration. Combined effect types of multiple stressors range from single-stressor dominance to complex interactions. Identifying prevalent combined effect types is critical for environmental management, as it helps to prioritise key stressors for mitigation. However, it remains unclear whether observed single and combined stressor effects reflect true ecological processes unbiased by sample size and length of stressor gradients. Therefore, we examined the role of sample size and stressor gradient lengths in 158 paired-stressor response cases with over 120,000 samples from rivers, lakes, transitional and marine ecosystems around the world. For each case, we split the overall stressor gradient into two partial gradients (lower and upper) and investigated associated changes in single and combined stressor effects. Sample size influenced the identified combined effect types, and stressor interactions were less likely for cases with fewer samples. After splitting gradients, 40 % of cases showed a change in combined effect type, 30 % no change, and 31 % showed a loss in stressor effects. These findings suggest that identified combined effect types may often be statistical artefacts rather than representing ecological processes. In 58 % of cases, we observed changes in stressor effect directions after the gradient split, suggesting unimodal stressor effects. In general, such non-linear responses were more pronounced for organisms at higher trophic levels. We conclude that observed multiple stressor effects are not solely determined by ecological processes, but also strongly depend on sampling design. Observed effects are likely to change when sample size and/or gradient length are modified. Our study highlights the need for improved monitoring programmes with sufficient sample size and stressor gradient coverage. Our findings emphasize the importance of adaptive management, as stress reduction measures or further ecosystem degradation may change multiple stressor-effect relationships, which will then require associated changes in management strategies.