Digestive Endoscopy submission (v10-DEN)

Article type: Original Article Target journal: Digestive Endoscopy Framing: Retrospective before–after observational study Spelling: US (Merriam-Webster) · References: Chicago Note style

Compression target: ≤ 3,000 words including abstract, excluding references, tables, figures. Detailed sensitivity analyses, BMI sub-structure detail, extended limitations, and the analogue-clock anecdote are relocated to the Supplement (unlimited under DEN rules).

Title

A Wall-Mounted Stopwatch and Propofol Dosing Patterns During Sedation Endoscopy: A Retrospective Before–After Observational Study

Running title: Stopwatch and propofol dosing in endoscopy

Author: Yong Bae Kim, MD¹

¹ Department of Endoscopy, Korea Association of Health Promotion (KAHP) Jeju Branch, 111 Yeonbuk-ro, Jeju-si 63136, Republic of Korea. ORCID 0000-0002-9933-438X.

Corresponding author: Yong Bae Kim, MD. Email: kcchgs@gmail.com; tel +82-(to add).

Preprint: A version of this work is posted on medRxiv (DOI 10.1101/2023.07.21.23292749).

Abstract (structured; target ≤ 250 words)

Objectives. Propofol titration during sedation endoscopy depends on judging elapsed time after each bolus, but subjective time perception compresses under cognitive load. We examined whether installing a large wall-mounted stopwatch in the endoscopist’s visual field was associated with changes in propofol dosing patterns, and whether any change differed by sex and body habitus.

Methods. Retrospective before–after analysis of routinely collected sedative esophagogastroduodenoscopy data from a single screening center performed by one endoscopist (n = 1,371; baseline 526, post-installation 845). The main process measure was weight-adjusted propofol dose (WAPD = total dose/body weight). Weight-adjusted doses were compared within each sex (Mann–Whitney), the total-dose distribution was compared across 10-mg levels, and a multivariable linear regression (robust SE) estimated how the female–male WAPD difference changed between periods, adjusting for age, weight, height, and procedure time.

Results. Median WAPD fell in women (1.379→1.288 mg/kg; p = 0.001) and rose non-significantly in men (1.235→1.264 mg/kg; p = 0.095). The female median total dose fell from 80 to 70 mg and the male median rose from 80 to 90 mg. The female–male WAPD difference narrowed from +0.145 to +0.023 mg/kg (adjusted period × sex interaction −0.097 mg/kg, 95% CI −0.147 to −0.048; p < 0.001), concentrated in the obese stratum. No severe adverse events were reported.

Conclusions. Stopwatch installation was associated with a redistribution of propofol dosing across sex and body habitus rather than a uniform dose change, consistent with a human-factors mechanism. Prospective audits recording per-bolus timing and sedation depth are warranted.

Keywords: cognitive aids; conscious sedation; endoscopy; human factors; propofol.

Introduction

Sedative esophagogastroduodenoscopy (EGD) is among the most common outpatient procedures in countries with screening programs, and propofol monotherapy is the default regimen for its rapid onset and clean recovery.¹ The same kinetics make it unforgiving: the therapeutic window is narrow in both dose and time. Redosing before the previous bolus peaks can cause cumulative oversedation; redosing too late can fail to prevent arousal. In a single-operator workflow both judgments — depth and elapsed time — are made repeatedly under cognitive load.

Subjective time perception compresses under task load and arousal,²,³ and acute-care human-factors research documents systematic underestimation of elapsed time. Cognitive aids that externalize information into the environment reduce predictable judgment errors under load, as shown for the surgical safety checklist⁴ and a systematic review of cognitive aids in emergencies.⁵ A large, visible digital stopwatch externalizes elapsed time: it is salient in the operator’s visual field and read rather than reconstructed from memory.

During the unit’s 2019 transition from midazolam–propofol to propofol monotherapy, an early case produced unexpectedly deep sedation after a bolus was given before the preceding dose had peaked; elapsed time at the bedside was hard to track from the analogue wall clock alone (detailed in the Supplement). This motivated installing a stopwatch to externalize elapsed time during titration.

We examined whether stopwatch installation was associated with measurable changes in propofol dosing during routine sedative EGD. During evaluation we observed that dosing changes differed by sex and body habitus, and centered the analysis on these patterns. These sex- and body-habitus analyses are post-hoc and hypothesis-generating.

Methods

Setting. A single screening center in the Republic of Korea. One board-certified endoscopist performs ~8,000 sedative upper endoscopies per year and decides every propofol dose; a procedural nurse administers the drug. No dedicated anesthesia provider is present. All cases are outpatient screening or diagnostic procedures in ASA I–II adults; biopsy cases were not performed in this unit during the study period. Pulse oximetry and supplemental oxygen are continuously available.

Intervention. A large digital LED wall clock with stopwatch function was mounted in the endoscopist’s direct visual field in June 2019 (Figure 1). It was started immediately after the initial propofol bolus and used to support a structured pause before reassessing the patient and deciding on additional boluses. Only the choice architecture around redosing changed; the dosing protocol — initial 0.5–1.2 mg/kg by age, each additional bolus 10 or 20 mg — was unchanged. No staff training, dose limits, or interval rule was introduced. Equipment cost was ~USD 30.

Design and data. Retrospective before–after analysis of routinely collected sedation EGD data comparing the baseline period (January–May 2019) with the post-installation period (July–October 2019). June 2019, the installation month, was excluded. All consecutive cases by the operator in each period were eligible; biopsy cases were excluded because biopsy lengthens procedure time and may increase total dose. Data were extracted from the routine clinical record.

Measures and analysis. The main process measure was weight-adjusted propofol dose (WAPD = total dose [mg] / body weight [kg]). Procedure time was the interval between first and last endoscopic image timestamps. We compared periods three ways: (i) within-sex WAPD (Mann–Whitney); (ii) the proportion of patients at each common total-dose level (70/80/90/100 mg; two-proportion z-test); (iii) the female–male median WAPD difference per period (stratified bootstrap 95% CI). To estimate how that difference changed between periods, we fitted a linear regression of WAPD on period, sex, their interaction, and age, weight, height, and procedure time, with robust (HC3) standard errors; the period × sex term gives the change in the gap. BMI was grouped as underweight (<18.5), normal (18.5–24.9), and obese (≥25 kg/m², KSSO/WHO Asian-Pacific threshold). Sensitivity analyses (propensity-score matching, January exclusion, alternative covariate sets, quantile regression, interrupted time-series) are in the Supplement. Analyses used Python 3.12 (statsmodels 0.14.6); two-sided p < 0.05 was significant.

Ethics. Approved by the Public Institutional Review Board designated by the Korean Ministry of Health and Welfare (P01-202102-11-001), with waiver of informed consent for retrospective de-identified analysis.

Results

Cohort. Of 1,373 records (526 baseline, 847 post-installation), two post-installation cases were excluded for missing procedure time, leaving 1,371 (baseline 526: 237 male, 289 female; post-installation 845: 405 male, 440 female; Table 1). Most characteristics were balanced; height was modestly higher (165.1 vs 163.9 cm; p = 0.020) and procedure time shorter (median 74 vs 78 s; p < 0.001) after installation. Unstratified median total dose and WAPD did not differ.

Sex-stratified WAPD (Table 2). Median WAPD moved in opposite directions: women 1.379→1.288 mg/kg (p = 0.001); men 1.235→1.264 mg/kg (p = 0.095).

Total-dose distribution (Figure 2). The female median total dose fell one 10-mg step (80→70 mg) and the male median rose one step (80→90 mg). Period differences in the dose distribution were significant in both sexes (women p < 10⁻⁵; men p = 0.003). Level-by-level proportions with z-tests are in Supplement Table S0.

Female–male difference. The median female–male WAPD difference fell from +0.145 mg/kg (95% CI +0.078 to +0.203) to +0.023 mg/kg (95% CI −0.010 to +0.066). In multivariable regression the period × sex interaction was −0.097 mg/kg (95% CI −0.147 to −0.048; p < 0.001), indicating a narrowed gap after installation. The estimate was preserved across all sensitivity analyses (range −0.085 to −0.125; Supplement).

BMI substructure (Table 3, Figure 4). The shrinkage concentrated in the obese stratum (≥25), where the female–male difference reversed from +0.068 to −0.071 (Δ −0.139), driven by a female reduction (median ΔWAPD −0.073; p = 0.043) and a male increase at trend level (+0.066; p = 0.096). The normal stratum showed a smaller reduction (+0.095→+0.031). The female reduction was also significant in the normal stratum (ΔWAPD −0.050; p = 0.001). The underweight stratum had only 4 male patients per period and is descriptive only.

Temporal trend (Figure 3). Median WAPD was already declining in both sexes before the intervention (monthly −0.038 mg/kg male, p = 0.008; −0.059 female, p < 0.001), with a level shift at July 2019. The female–male gap was stable across pre-intervention months (slope p = 0.51), so its abrupt narrowing at July is not a continuation of trend. An interrupted time-series adjustment (Supplement) attenuated the interaction toward null, limited by few monthly points.

Adverse events. No severe adverse events (resuscitation, oxygen escalation beyond standard supplementation, or aborted procedures) were reported in either period. Structured capture of oxygen saturation, recovery time, sedation-depth scores, and minor airway events was not available, so safety statements are narrative.

Discussion

In this single-center, single-operator before–after analysis, stopwatch installation was associated with a redistribution of propofol dosing across sex and body habitus rather than a uniform dose reduction: the female median total dose fell one 10-mg step, the male median rose one step, and the female–male WAPD gap narrowed from +0.145 to +0.023 mg/kg.

The change concentrated in the obese stratum, where the gap reversed sign. This aligns with bariatric pharmacokinetics: total-body-weight scaling risks over-titration in obesity because clearance tracks lean body weight rather than total weight.¹⁴,¹⁵ Obese women, with lower lean-mass fraction than obese men, are especially exposed to weight-based over-dosing under time compression; an externalized time cue plausibly suppressed unnecessary redosing in obese women while licensing adequate loading in obese men. A female reduction was also seen in the normal-BMI stratum, consistent with the recognized earlier propofol arousal in women,⁶⁻¹³ so we avoid a purely low-BMI reading.

This convergence should not be read as evidence that identical weight-normalized dosing is optimal across sex or habitus. WAPD is mechanically influenced by body weight, and the cross-sectional female–male WAPD difference within each BMI category (ignoring period) was small, indicating that much of the overall sex difference reflects body-size distribution. The within-stratum gap nonetheless shifted between periods — most clearly reversing in the obese stratum — so the pattern was not explained by BMI redistribution alone.

The male total-dose increase (80→90 mg) was concentrated at higher dose levels rather than spread across all men, and in both periods obese men received lower mg/kg than normal-BMI men — preserving expected pharmacology.¹⁴,¹⁵ This is consistent with the time cue reducing cautious under-dosing rather than causing generalized oversedation. We do not claim the post-intervention pattern is optimal, only that the pre-existing sex gap narrowed.

To our knowledge this is the first study to examine whether a low-cost environmental device — distinct from a checklist or protocol — is associated with shifted dosing during routine, high-volume procedural sedation, where the judgment under load is a recurrent timing decision rather than a discrete event.

Limitations. The before–after design lacks a concurrent control, so we describe the intervention as associated with, not causing, dosing change. Median WAPD was already declining before the intervention, and an interrupted time-series adjustment attenuated the interaction toward null; however, the female–male gap was stable pre-intervention, so its level shift is not pre-trended in the way absolute levels are. Weight-based scaling is an imperfect instrument, so the gap reduction is partly a redistribution along the BMI gradient correlated with sex, plus a within-stratum component (clearest in the obese stratum). The sex/BMI findings are post-hoc and hypothesis-generating. No structured safety, sedation-depth, or per-bolus timing data were captured, so the proposed mechanism is inferential and no safety claim is made. The single-operator design aids internal consistency but limits generalizability. Further limitations and unmeasured confounders are detailed in the Supplement.

Conclusions

Stopwatch installation was associated with a redistribution of propofol dosing patterns: the female median total dose fell one 10-mg step, the male median rose one step, and the female–male weight-adjusted gap narrowed substantially, most clearly in obese patients. Clinically this represents more uniform propofol exposure across sex and body habitus, achievable with a single low-cost device. Prospective audits recording per-bolus timing, sedation depth, and structured safety outcomes are needed to test the proposed mechanism.

Acknowledgments

The wall-stopwatch line-drawing in Figure 1 was created by [artist name, to be inserted], used with permission. The author used a generative AI assistant (Anthropic Claude) for language editing, drafting support, and manuscript formatting; the author conceived the study, performed and verified all data analysis, figures, and citations, and takes full responsibility for the content. No AI tool is listed as an author.

Conflict of Interests

The author declares no conflict of interests for this article.

Funding Information

None.

Ethics Statement

Approval of the research protocol: Public Institutional Review Board designated by the Korean Ministry of Health and Welfare, approval no. P01-202102-11-001. Informed Consent: N/A (waiver granted for retrospective de-identified analysis). Registry and Registration No.: N/A (retrospective study). Animal Studies: N/A.

References (Chicago Note style — numbered as they occur)

1. V. Wadhwa, D. Issa, S. Garg, R. Lopez, M. R. Sanaka, and J. J. Vargo, “Sedation and Anesthesia in GI Endoscopy in 2025: How, Who, and Why,” Gastrointestinal Endoscopy (2025), https://doi.org/10.1016/j.gie.2025.09.014.

2. R. A. Block, P. A. Hancock, and D. Zakay, “How Cognitive Load Affects Duration Judgments: A Meta-Analytic Review,” Acta Psychologica 134, no. 3 (2010): 330–343, https://doi.org/10.1016/j.actpsy.2010.03.006.

3. I. Polti, B. Martin, and V. van Wassenhove, “The Effect of Attention and Working Memory on the Estimation of Elapsed Time,” Scientific Reports 8 (2018): 6690, https://doi.org/10.1038/s41598-018-25119-y.

4. A. B. Haynes, T. G. Weiser, W. R. Berry, et al., “A Surgical Safety Checklist to Reduce Morbidity and Mortality in a Global Population,” New England Journal of Medicine 360, no. 5 (2009): 491–499, https://doi.org/10.1056/NEJMsa0810119.

5. P. R. Greig, D. Zolger, D. N. Onwochei, N. Thurley, H. Higham, and N. Desai, “Cognitive Aids in the Management of Clinical Emergencies: A Systematic Review,” Anaesthesia 78, no. 3 (2023): 343–355, https://doi.org/10.1111/anae.15939.

6. S. C. Hoymork and J. Raeder, “Why Do Women Wake Up Faster Than Men From Propofol Anaesthesia?,” British Journal of Anaesthesia 95, no. 5 (2005): 627–633, https://doi.org/10.1093/bja/aei245.

7. I. Loryan, M. Lindqvist, I. Johansson, et al., “Influence of Sex on Propofol Metabolism, a Pilot Study,” European Journal of Clinical Pharmacology 68, no. 4 (2012): 397–406, https://doi.org/10.1007/s00228-011-1132-2.

8. E. Choong, I. Loryan, M. Lindqvist, et al., “Sex Difference in Formation of Propofol Metabolites: A Replication Study,” Basic & Clinical Pharmacology & Toxicology 113, no. 2 (2013): 126–131, https://doi.org/10.1111/bcpt.12070.

9. S. Maeda, Y. Tomoyasu, H. Higuchi, Y. Honda, M. Ishii-Maruhama, and T. Miyawaki, “Female Patients Require a Higher Propofol Infusion Rate for Sedation,” Anesthesia Progress 63, no. 2 (2016): 67–70, https://doi.org/10.2344/0003-3006-63.2.67.

10. H. Pleym, O. Spigset, E. D. Kharasch, and O. Dale, “Gender Differences in Drug Effects: Implications for Anesthesiologists,” Acta Anaesthesiologica Scandinavica 47, no. 3 (2003): 241–259, https://doi.org/10.1034/j.1399-6576.2003.00036.x.

11. M. M. Sahinovic, M. M. R. F. Struys, and A. R. Absalom, “Clinical Pharmacokinetics and Pharmacodynamics of Propofol,” Clinical Pharmacokinetics 57, no. 12 (2018): 1539–1558, https://doi.org/10.1007/s40262-018-0672-3.

12. R. H. Thaler and C. R. Sunstein, Nudge: Improving Decisions About Health, Wealth, and Happiness (Yale University Press, 2008).

13. A. Gawande, The Checklist Manifesto: How to Get Things Right (Metropolitan Books, 2009).

14. F. Servin, R. Farinotti, J. P. Haberer, and J. M. Desmonts, “Propofol Infusion for Maintenance of Anesthesia in Morbidly Obese Patients Receiving Nitrous Oxide,” Anesthesiology 78, no. 4 (1993): 657–665, https://doi.org/10.1097/00000542-199304000-00008.

15. J. Ingrande and H. J. M. Lemmens, “Dose Adjustment of Anaesthetics in the Morbidly Obese,” British Journal of Anaesthesia 105, suppl. 1 (2010): i16–i23, https://doi.org/10.1093/bja/aeq312.

NOTE: citation superscripts in the text currently follow the v9 numbering (Nudge/Checklist were [4]/[5], arousal cluster [8–12], PK [13–15]). Reference numbering above has been re-sequenced to order of first appearance per Chicago Note; in-text superscripts must be re-mapped to match this list before submission (flagged for the editing pass).

Tables (unchanged from v9 content; vertical lines to be removed, footnote symbols †‡§¶ for DEN)

Table 1. Patient and procedural characteristics by period (n = 1,371)

SD, standard deviation; IQR, interquartile range; BMI, body mass index; WAPD, weight-adjusted propofol dose. Continuous variables by t-test or Mann–Whitney; categorical by chi-square.

Table 2. Sex-stratified weight-adjusted propofol dose (n = 1,371)

WAPD, weight-adjusted propofol dose; IQR, interquartile range. Median [IQR]; WAPD is right-skewed in all four groups. Level-by-level proportions in Supplement Table S0.

Table 3. Female–male WAPD difference by BMI category and period

KSSO/WHO Asian-Pacific obesity threshold (≥25). Underweight male cells (n = 4) descriptive only. Within-sex Mann–Whitney details in Supplement Table S-BMI.

Figure legends

Figure 1. Line-drawing illustration of the wall-mounted digital LED stopwatch (elapsed time 01:46) in the endoscopist’s direct visual field, beside an analogue reference clock; the endoscopy monitor is at left. The stopwatch is started after the initial propofol bolus. Illustration by [artist name], used with permission. A blurred clinical photograph of the same scene is Supplementary Figure S3.

Figure 2. Distribution of total propofol dose by sex and period (n = 1,371). Proportion of patients at each discrete total-dose level, Before and After overlaid. Female median 80→70 mg (p < 10⁻⁵); male median 80→90 mg (p = 0.003). Level-by-level z-tests in Supplement Table S0.

Figure 3. Monthly run chart of weight-adjusted propofol dose (Before: Jan–May 2019; After: Jul–Oct 2019). (A) Monthly median WAPD by sex; both decline before the intervention. (B) Monthly female–male WAPD difference; stable pre-intervention and dropping at the boundary.

Figure 4. BMI substructure of stopwatch-associated dosing change (KSSO/WHO: underweight <18.5, normal 18.5–24.9, obese ≥25). (A) Female–male median WAPD difference by BMI category × period; the difference shrinks across categories and reverses sign only in the obese stratum. (B) Within-sex change in median WAPD (After − Before) with bootstrap 95% CIs; female (upper) and male (lower) separated by shading. Markers: *p < 0.05; †p < 0.10.