Number of retractions

A total of 348,117 publications have been produced with support from Indian-funding agencies from 2000 to 2024. 368 papers were retracted in the total publications of Indian-funded research between 2000 and 2024. Among 368 papers, 344 were Articles, 18 Review Articles, 5 Conference papers, and 1 Letter to the Editor.

Figure 1 illustrates the trajectory of publication and retraction trends in Indian-funded research over the years. Between 2000 and 2007, a total of 255 papers were published, with no retractions observed during this period. The first instance of retraction appeared in 2008, with four papers. From that point onward, retractions steadily increased, reaching their peak around 2017. We witness a gradual decline from 2018 onward. This trend suggests a growing awareness and enforcement of research integrity practices over time by the publishers (Collaborative Working Group from the conference “Keeping the Pool Clean: Prevention & Management of Misconduct Related Retractions, 2018).

Fig. 1

Chronological Trend in the publications and retractions in Indian-funded research (2000–2024)

Reasons for retraction

Retractions are rarely the result of a single reason and often overlap, and the language used in notices varies in clarity and transparency. A precise classification of retraction reasons remains elusive (P. Sharma et al., 2023). Retraction Watch lists 111 reasons for retracting articles (Retraction Watch, n.d.), which we refer to and follow in this study for retraction reasons analysis. Table 1 highlights the 55 reasons for the retraction of 368 research articles in Indian-funded research. Image-related issues are significant contributors to retractions, with the following specific reasons; Duplication of images (36.04%), concerns or issues about images (16.53%), manipulation of images (11.92%), errors in images (8.40%), falsification or fabrication of images (2.44%), plagiarism of images (1.90%), and unreliable images (0.54%) (see Table 1). Altogether, these issues account for approximately 53% of all retractions due to image misuse or misrepresentation (see Table 2). In addition to image-related problems, issues concerning the generation, analysis, or interpretation of raw data are another critical factor, as 40% of retractions stem from concerns about data integrity and result validity. Systemic weaknesses in editorial and ethical oversight contribute to 20% of retractions. Furthermore, 18% of retractions are related to author-related issues.

Table 1 Frequency counts and percentage of retraction reasonsTable 2 The broad categorization of retraction reasonsWhat are the citation trends before and after retraction?

There were 13,114 total citations to 368 retracted papers. Figure 2 illustrates pre- and post-retraction citations between 2008 and 2024. It also highlights the year-wise citation trends, which indicate a downward trend in citations to retracted papers in recent years, at least after 2017. Citation counts for each paper before and after its retraction were collected and analyzed. The analysis revealed that a considerable volume of 30.32% (3977 citations) were made after the retraction. This pattern may indicate a lack of awareness regarding retractions, inadequate communication within the academic community, or ignorance by researchers about retracted papers (Schneider et al., 2022).

Fig. 2

A stacked bar chart illustrates pre- and post-retraction citations for each year. The overall height of each bar represents the total number of citations, while the color-coded segments differentiate between pre- and post-retraction citations

Do journal impact factor and quartile rankings influence citation patterns?

To examine whether journal characteristics influence citation patterns before and after retraction, we analyzed the distribution of retracted papers across journals and their characteristics, in terms of impact factor and quartile ranking. A total of 368 retracted papers were published in 208 journals, many of which are recognized for their high visibility and citation impact. High volume of retracted papers was published in Q1 journals with 170 papers (46.20%), followed by Q2, 121 papers (32.88%), Q3, 49 papers (13.32%), and Q4, 14 papers (3.80%). Prominent journals such as PLOS ONE (28 papers), Bioresource Technology (24 papers), Journal of Biological Chemistry (19 papers), and RSC Advances (15 papers) published a major proportion of retracted papers. The high volume of retracted papers in high-impact and Q1 journals indicates that their wide and critical readership contributes to a greater potential of error detection and subsequent corrections.

In terms of citation dynamics, 69.67% of citations were pre retraction citations across journal quartile ranks. The analysis of post-retraction citation patterns across journal quartiles revealed that Q2 journals recorded the highest proportion (37.48%) of post-retraction citations (among 3191 total citations in Q2 journals 1995 are post retraction citations). Q4 journals followed with 35.66% post-retraction citations (among 143 total citations, 92 post-retraction), and Q3 journals showed a similar pattern with 35.23% post-retraction citations (among 1422 total citations, 921 post-retraction). Q1 journals accounted for the lowest proportion of post-retraction citations (26.55%), despite having a large volume of citations (among 8074 total citations, 5930 post-retraction). These findings suggest that comparatively high-impact journals are more effective at communicating retractions to their readers, or that authors citing these journals are more likely to adhere to research integrity standards.

Table 3 The top 20 journals with high retractionsIs there an influence of self-citations on retracted papers?

Figure 3 shows the concentration of self-citations and general citations to the retracted papers between 2008 and 2024. Among the total 13,114 citations, the majority (11,974) are general citations not coming from the same author group but from other sources. The author groups associated with the retracted papers cited their own retracted works (self-cited) 1140 times across 230 papers, accounting for 8.69% of total citations. Overall, the average number of self-citations was 3.10.

Fig. 3

Self-Citations vs General Citations in retracted papers

An interesting pattern was observed within the self-citations. About 152 self-citations were made post-retraction. This indicates that retracted papers continue to get self-cited, giving them undue visibility despite their invalidation.

Further, we identified 168 retracted citing papers among the total citations. Of them, 110 had self-citations (64.48%). About 1.28% of citing papers had a cascading retraction effect. Both self-citations in original retracted papers and the tendency of citing retracted papers have contributed to this cascading retraction effect.

How do citation patterns vary before and after retraction?

As retrieved from scite.ai, there were 14,242 citations for the 368 retracted papers. Among these, 10,193 were pre-retraction citations and 3947 were post-retraction citations.

Contextual citation patterns

Figure 4 shows the context of citing papers classified into supporting, mentioning, and contrasting citations before and after retraction. In both pre-retraction and post-retraction, mentioning citations were dominant (3922 and 1206, respectively), indicating the neutral approach of citing authors. The actual citation polarity can be assessed by analyzing supporting citations and contrasting citations. A smaller proportion of 258 and 67 supporting citations were found pre and post retraction, respectively. Contrasting citations were the least in both pre and post retraction (16 and 8, respectively). This pattern indicates a clear lack of critical engagement with scholarly works. The instances of both supporting and contrasting citations reduced post retraction. However, the overall pattern of contextual citations remains the same pre and post retraction.

Fig. 4

Contextual Citation Patterns

Sectional citation patterns

Understanding sectional citation patterns is important to comprehend the level of maturity of scholarly works and their utility for furthering the scholarly discourse.

Figure 5 presents the distribution of citations to retracted papers across different sections of the citing papers. A large proportion of citations appear in the introduction Sect. (1494 before and 419 after retraction), indicating that the authors of these papers most frequently reference them to establish background or research context. Citations in the discussion section accounted for 815 before retraction and 254 after retraction, showing that many authors continue referencing these works when comparing their own findings with previous studies. Citations in the results section generally appear to support specific findings. 117 such citations were found after retraction compared to 389 citations before retraction. Ideally, such citations should stop after retraction, but instead, they continued occurring. Citations in the methods Sect. (199 before and 47 after retraction) indicate that authors frequently use retracted works as methodological references or procedural frameworks. This indicates that, in spite of retraction, certain methodological approaches were considered valuable for subsequent research, reflecting the continued influence of technical aspects of retracted literature in scientific practice. There is a clear reduction in citation counts in each section post-retraction. However, continued citing of retracted papers poses serious concerns about the validity of citing papers.

Fig. 5

Sectional Citation Patterns

We present three detailed cases focusing on the selected highly cited papers within our dataset to illustrate citation patterns.

Illustration-1

Sudheesh Kumar, P. T., Lakshmanan, V.-K., Anilkumar, T. V., Ramya, C., Reshmi, P., Unnikrishnan, A. G., Nair, S. V., & Jayakumar, R. (2012). RETRACTED: Flexible and Microporous Chitosan Hydrogel/Nano ZnO Composite Bandages for Wound Dressing: In Vitro and In Vivo Evaluation. ACS Applied Materials & Interfaces, 4(5), 2618–2629. https://doi.org/10.1021/am300292v.

This article was published in 2012 in ACS Applied Materials & Interfaces and retracted in 2019 due to concerns regarding “duplication and inappropriate presentation of images,” which were significant enough to affect the main conclusions of the study. According to Scite.ai, the article has been cited825 times since its publication. Citations continued even after the retraction (354 citations), with 53 citations in 2023, 50 citations in 2024, and 36 citations in 2025 (as of data retrieval in October 2025), indicating an ongoing post-retraction influence of the article (see Fig. 6).

Fig. 6

Number of Citations per year-Sudheesh Kumar, P. T. et al. article

The analysis of contextual citations indicates that the majority of citations fall under the "mentioning" category, with no critical engagement (see Fig. 7a). Despite the retraction being issued due to image duplication and misrepresentation that could impact the study’s conclusions, four articles cited the work in a supportive context. Interestingly, three of these citations occurred after the retraction. Additionally, most articles referenced this paper either in their introduction to provide background or establish context or within the results section, potentially using it to compare or support their own findings (see Fig. 7b).

Fig. 7Illustration-2

Singhania, R. R., Patel, A. K., Sukumaran, R. K., Larroche, C., & Pandey, A. (2013). RETRACTED: Role and significance of beta-glucosidases in the hydrolysis of cellulose for bioethanol production. Bioresource Technology, 127, 500–507. https://doi.org/10.1016/j.biortech.2012.09.012.

The co-author of this review paper, Prof. Ashok Pandey, affiliated with the Council of Scientific and Industrial Research (India), was the editor-in-chief of the journal Bioresource Technology, in which as many as 43 of his papers were retracted by the publisher Elsevier (Prasad, 2024). According to the journal’s official website, these retractions followed “a journal-wide investigation identified violations of the journal’s policies on authorship and conflict of interest related to the submission and peer-review process.

This review article, published in 2013 and retracted in 2024, had accumulated 628 citations up to the time of analysis (see Fig. 8). 21 of these citations occurred after the retraction, indicating the paper continued to influence scholarly work despite its formal withdrawal from the scientific record.

Fig. 8

Number of Citations per year-Singhania, R. R. et al. article

The article received a total of 8 supporting citations, 313 mentioning citations, and no contrasting citations, according to Scite.ai’s classification. Despite its retraction, the article continues to be cited likely because its findings were considered scientifically valid and remain relevant within the field. The continued citation may be influenced by the prominence of the lead author, who holds significant recognition in the Indian biotechnology research community. This suggests that an author’s reputation can play a significant role in the persistence of citations, even after formal retraction.

Illustration-3

Naqvi, S., Samim, M., Abdin, M., Ahmed, F. J., Maitra, A., Prashant, C., & Dinda, A. K. (2010). Concentration-dependent toxicity of iron oxide nanoparticles mediated by increased oxidative stress. International journal of nanomedicine, 5, 983–989. https://doi.org/10.2147/IJN.S13244 (Retraction published Int J Nanomedicine. 2022 Mar 25;17:1459–1460. https://doi.org/10.2147/IJN.S367448.)

This article was published in 2010 and retracted in 2022 after 12 years due to the “duplication of features within the TEM image, " a concern that raised questions about the integrity of the data presented. The authors were unable to provide a satisfactory explanation for the alleged duplication or produce the original image data. The retraction notice was prominently displayed on each page of the article with a digital watermark stating “Retracted”. Despite its retraction, the article has continued to influence the scientific community, having received a total of 522 citations (Fig. 9), including 84 citations after the retraction, demonstrating ongoing academic engagement with the work, either knowingly or unknowingly.

Fig. 9

Number of Citations per year-Naqvi, S et al. article

The article received recognition within the scholarly community by accumulating 18 supporting citations, 268 mentioning citations, and no contrasting citations. Even after its retraction in 2022, the article continued to be cited in a supportive context, with three such citations appearing post-retraction. In these instances, the authors of the citing papers referenced the retracted study within their results and discussion sections, indicating continued reliance on its findings despite its compromised status.

Altmetric attention score (AAS) of retracted papers

To further explore the broader impact of retracted publications in Indian-funded research, we analyzed their AAS retrieved from altmetric.com. This analysis highlighted diverse levels of engagement across multiple platforms, reflecting the extent to which the retracted studies attract attention beyond traditional academic citations (Table 4).

Table 4 Altmetric Attention Score of retracted publicationsSocial media mentions

A detailed review of the associated social media mentions, including posts on X (formerly Twitter), blog mentions, and news articles, reveals that these mentions primarily serve to highlight or criticize the retractions. A significant portion of blog mentions (111 mentions across 89 articles) originates from two prominent sources, i.e., “For Better Science” (https://forbetterscience.com/) and “Retraction Watch” (https://retractionwatch.com/). These blogs are known for critically reporting on research misconduct and have played a key role in exposing the reasons behind the retractions through detailed investigative reporting.

Patent mentions

Interestingly, 19 publications were cited in 118 patent mentions, indicating the influence of these papers on innovation. To better understand the nature of the patent mentions, we conducted a focused case analysis of two highly cited (patent mentions) retracted papers.

AAS illustration-1

Naqvi, S., Samim, M., Abdin, M., Ahmed, F. J., Maitra, A., Prashant, C., & Dinda, A. K. (2010). Concentration-dependent toxicity of iron oxide nanoparticles mediated by increased oxidative stress. International journal of nanomedicine, 5, 983–989. https://doi.org/10.2147/IJN.S13244 (Retraction published Int J Nanomedicine. 2022 Mar 25;17:1459–1460. https://doi.org/10.2147/IJN.S367448.)

This article, published in 2010, explored the toxicity of iron oxide nanoparticles and their biomedical implications, an area of intense interest in drug delivery, imaging, and cancer therapy. The study was retracted 12 years later in 2022, with the reason cited as "duplication of features within the TEM image", which raised concerns about the integrity and reproducibility of the data. Despite the retraction, the article had already left a substantial footprint in patent references. It was cited in 66 patent documents, of which 35 citations occurred after the retraction. This post-retraction citation pattern indicates that the study’s conceptual framework or conclusions may still be considered valuable by citing authors.

AAS illustration-2

Jayandharan, G. R., Aslanidi, G., Martino, A. T., Jahn, S. C., Perrin, G. Q., Herzog, R. W., & Srivastava, A. (2011). RETRACTED: Activation of the NF-κB pathway by adeno-associated virus (AAV) vectors and its implications in immune response and gene therapy. Proceedings of the National Academy of Sciences, 108(9), 3743–3748. https://doi.org/10.1073/pnas.1012753108.

This study focused on the immune response triggered by AAV vectors, viral tools extensively used in gene therapy. It was retracted in 2017, six years after publication, due to multiple issues, including data duplication, image manipulation, and data fabrication/falsification. However, the journal’s retraction notice stated that "none of the major conclusions reached in the paper are affected" by these concerns. Despite the formal retraction, the paper was cited in 24 patents, with 16 of those patent citations occurring after the retraction. This suggests that the study’s core insights into immune activation and viral vector design continued to inform innovation in gene therapy, likely because the findings aligned with or were validated by subsequent independent studies.

Peer review mentions

15 retracted papers received 64 peer review mentions on PubPeer.com. PubPeer is a scientific forum or journal club where scientific publications are discussed after publication (Ortega, 2022) and also allow users to critique published papers, often anonymously, and has become a platform for identifying potential issues or misconduct in scientific research.

Policy mentions

6 retracted papers were found to be cited in policy documents issued by prominent international organizations, such as The World Health Organization (WHO), which cited 4 retracted papers, The International Renewable Energy Agency (IRENA) cited 1, The National Academies Press (U.S.) included 1 mention, and The Publications Office of the European Union referenced one.