Addressing methodological and analytical pitfalls on medical devices related adverse events

Sir,

We read with interest the recent article by Veeramani et al,¹ published in June 2025 issue of Indian Journal of Medical Research. The authors address the critical topic of materiovigilance, and we appreciate their effort in identifying patterns of medical device-related adverse events (MDAEs) in an ICU setting. However, we would like to point out certain issues which we feel detract from the methodological soundness of the study and affect the reliability of the findings.

First, the study is described as cross-sectional, which typically involves data collection at a single point or over a narrow window of time.² The authors, however, state that individuals were monitored... daily for new adverse events, in the abstract.¹ This description is characteristic of a prospective, longitudinal study. Further, the authors do not provide a justification for their sample size of 90 participants, for example, by assessing the volume of prior hospital records or a pilot period, nor do they provide the details of their convenient sampling method. One does not get to know, for instance, whether healthcare providers other than doctors were involved in handling these devices and whether weekends or specific shifts were excluded. This impacts the study’s generalizability.

Key analytical choices require clarification. The authors provide operational definitions for causality assessment but do not cite a reference or standard guideline for the five-point scale. The Table III is titled Predictors of MDAEs, but the analysis involves comparison between different categorical variables with respect to their male female distribution rather than whether medical device-related adverse events occurred or not. The table title is therefore misleading, as the table has analysed the association of gender with various variables like comorbidities, length of stay, etc. and nothing else.³ In Figure 2, the categorisation of MDAE types is problematic; events like Ventilator-Associated Pneumonia (VAP) or Catheter-Associated Urinary Tract Infection (CAUTI) could arguably be classified under both Opportunistic infections and Procedural errors (e.g., lack of following proper aseptic practices).

The data presented in Table II ‘Incidence of Medical device causing adverse events’ has several lacunae: (i) the sample size (n) is not written. However, we infer that it may be the total study population i.e., 90 patients, (ii) the table should have had a footnote stating that an individual may have experienced more than one event, (iii) the table enlisted non-specific, expected outcomes e.g., Inability to speak with an endotracheal tube and confounding events e.g., bedsores, which is more likely a nursing-related issue, rather than a device-related one. The separate listing of events for Non-Invasive Ventilation (NIV), Bi-level Positive Airway Pressure (BiPAP), and Continuous Positive Airway Pressure (CPAP) is redundant, as the latter two are modes of the first. In the footnote to Table II, BiPAP is incorrectly expanded.

We believe that clarification of these methodological, analytical, and data-reporting issues is essential for the valid interpretation of the study and for guiding future materiovigilance research in India.⁴