The problem they describe is publication bias, a phenomenon documented since at least 1990, when researcher Kay Dickersin showed in JAMA that studies with favourable results are systematically more likely to be submitted and published than those reporting negative or inconclusive findings. The distortion operates in two ways: entire studies may vanish from the record, or favourable outcomes may be selectively emphasised within manuscripts that do appear. Subsequent systematic reviews have confirmed that the pattern extends across medicine, and analyses of the general and orthopaedic surgical literature have found positive findings outnumbering negative ones, raising concerns about how complete the evidence available to clinicians really is.

Abdominal wall surgery may be especially vulnerable. The specialty is marked by rapid technical innovation and a constant stream of new meshes, fixation systems, minimally invasive approaches and robotic platforms. Early favourable studies attract attention, funding and conference slots, while research showing no superiority, no clinically meaningful difference, or an unfavourable balance between benefits and harms risks fading into obscurity. Because hernia repair outcomes such as recurrence may take years to emerge, and recruitment into surgical trials is often difficult, the field is particularly susceptible to type II error, the false negative in which a real effect goes undetected simply because the study was too small or too short.

Part of the confusion, the authors argue, lies in language. The terms negative, neutral, null and non-significant are often used interchangeably, yet they describe distinct situations. A negative result generally means the findings do not support the original hypothesis or suggest the intervention performs worse than expected. A neutral or null result indicates that no statistically significant difference was detected between study groups. An inconclusive result means the evidence is too imprecise to determine whether a clinically important difference exists. Crucially, none of these findings automatically demonstrates equivalence. A superiority trial that fails to reject the null hypothesis has not proven that two treatments are the same; the absence of statistical significance may reflect a genuinely small effect, or it may stem from inadequate sample size, low event rates, imprecise measurement, incomplete adherence or insufficient follow-up.

This distinction carries real clinical weight. Claiming that two procedures are equivalent merely because a comparison produced a P value above 0.05 is a statistical error that equivalence and non-inferiority trials, with their dedicated designs and analyses, exist to avoid. The goal of publishing neutral research, the authors stress, is not to flood the literature with uninterpretable non-significant findings, but to ensure that well-designed studies are judged on the relevance of their question and their methodological quality rather than on whether they deliver a positive result.

Neutral results can be genuinely valuable. A reliable finding that two procedures achieve similar recurrence rates may be highly informative when one option is less costly, simpler to perform or faster in the operating room. Comparable outcomes can support a less invasive approach, challenge unnecessary technological investment, or justify dropping an operative step altogether. Neutral studies also define the limits of innovation: a new procedure does not need to be harmful to lack clinical value, and may simply fail to improve outcomes enough to compensate for greater complexity, cost, resource use or learning requirements. Surgeons and patients, the authors note, need to know not just whether a difference is statistically detectable, but whether its magnitude is large enough to matter.

Real examples from the field illustrate the point. In a multicentre randomised trial comparing robotic with laparoscopic ventral hernia repair, robotic repair did not reduce hospital stay during the first 90 postoperative days while requiring longer operating times and incurring higher costs; subsequent evaluation of patient-reported outcomes in the PROVE-IT trial found broadly comparable results between the approaches. The authors are careful to note that these studies do not imply robotic surgery lacks value in all settings, but they demonstrate that the anticipated advantages of the more advanced platform were not evident for the outcomes evaluated. Similarly, a randomised trial of laparoscopic inguinal hernia repair found no early recurrence benefit from mesh fixation, while fixation increased costs and was associated with more chronic pain, and a systematic review of randomised trials in open inguinal repair found no convincing differences in recurrence or infection between fixation strategies. When effectiveness is comparable, differences in pain, cost, operative time and technical complexity become the decisive factors.

Prophylactic mesh for parastomal hernia prevention offers perhaps the clearest case of evidence challenging initial optimism. The STOMAMESH randomised trial found that sublay mesh did not prevent parastomal hernia. Five-year follow-up from the PREVENT trial reported that prophylactic mesh was safe but did not reduce long-term incidence, and a pooled long-term analysis of cohorts from three randomised trials likewise concluded that prophylactic mesh did not prevent parastomal hernia over extended follow-up. A recent systematic review and meta-analysis of eight randomised trials with long-term follow-up appears to cast further doubt on the initial enthusiasm. Yet the authors caution against swinging to the opposite extreme: mesh configuration, position, operative technique, patient selection and outcome definition remain unresolved questions, and the findings should not be read as proof that every prophylactic mesh strategy is ineffective. Other studies reveal genuine trade-offs rather than clean wins or losses, with prophylactic intraperitoneal mesh potentially reducing incisional hernia while increasing early pain or wound-related morbidity, and lightweight mesh reducing chronic discomfort while raising recurrence concerns in selected settings.

The consequences of selective publication ripple far beyond individual studies. When positive findings are visible but neutral or unfavourable results remain unavailable, meta-analyses may overestimate treatment effects and underestimate uncertainty. Systematic reviews cannot incorporate studies they cannot locate, and guideline panels cannot assess unpublished clinical data, failed replications or long-term follow-up, so recommendations may appear stronger and more definitive than the total evidence justifies. Non-publication also creates research waste: clinical studies demand protocol development, ethical review, recruitment, data collection, follow-up and analysis, and much of that investment is lost when findings are deemed insufficiently exciting. Worse, other groups may inadvertently repeat interventions already shown to be ineffective, exposing additional patients to procedures that previous research had already failed to support. There is an ethical dimension too, since participants contribute to research expecting their involvement to generate knowledge that benefits future patients, an expectation that non-publication quietly betrays.

The authors propose a shared programme of reform. Researchers should register comparative studies and define their methodology before results are known, and submit completed studies regardless of whether the original hypothesis was supported. Authors should avoid describing non-significant findings as evidence of equivalence without considering statistical power and confidence intervals. Reviewers should evaluate the clinical question, risk of bias, precision and transparency before considering the direction of findings. Editors should explicitly welcome negative, neutral, replication and long-term follow-up studies, basing decisions on methodological validity rather than statistical significance or perceived novelty. Scientific societies should give non-positive research visibility at conferences and in consensus processes, funding bodies should require dissemination plans that apply regardless of outcome, and systematic reviewers should search trial registries, protocols and conference proceedings for completed but unpublished studies, reducing confidence in the evidence when selective publication is suspected. The EQUATOR Network’s reporting guidelines, the authors add, are particularly important for neutral studies because they allow readers to judge whether the results constitute credible evidence. Ultimately, they argue, the solution is not the unselective publication of every null finding but the dissemination of clinically relevant, methodologically rigorous and transparently reported research, because a specialty that publishes only positive findings cannot reliably assess the true value of its interventions.

Subject of Research: Publication bias and the value of negative and neutral results in abdominal wall surgery research

Article Title: Negative and neutral results matter in abdominal wall surgery

Article References: López-Cano, M., & Garcia-Alamino, J. M. (2026). Negative and neutral results matter in abdominal wall surgery. Journal of Abdominal Wall Surgery, 5, Article 17443. https://doi.org/10.3389/jaws.2026.17443

Image Credits: AI Generated

DOI: 10.3389/jaws.2026.17443

Keywords: abdominal wall surgery, publication bias, hernia repair, negative results, neutral results, randomised trials, mesh fixation, parastomal hernia, robotic surgery, evidence-based surgery, statistical significance, research transparency