Page 68 - 北京京煤集团总医院第十一届·2023学术年会论文集

北京京煤集团总医院 第十一届·2023 学术年会论文集

the risk factors for delayed bleeding after colorectal polypectomy; (2) duplicate publications
repeatedly; (3) animal experiments, reviews, case reports, conference abstracts, dissertations; (4)

studies without a control group; (5) the same researcher published similar documents; and (6)
documents with incomplete data, unclear description, unavailable data, and limited research group.
2.3. Literature screening and data extraction
All the studies were independently screened by two researchers (ZX, SL). The researchers screened

the titles, abstracts, and full texts in accordance with the inclusion and exclusion criteria, and the
data were extracted into an Excel spreadsheet. The extracted data included (1) general data,
including first author's name, year of publication, and type of study; (2) baseline data, including
sample size, sex, and age; (3) patient factors; (4) polyp factors, including polyp number, polyp

diameter, polyp shape, polyp location, and polyp pathological type; (5) operation factors; and (6)
the same effect size and 95% confidence interval (CI) were obtained by multivariate regression
analysis. Any disagreement was resolved through discussion, consultation with a third researcher
(JX) if necessary, and discussion or arbitration.

2.4. Quality assessment
Each study was independently evaluated by two researchers (ZX, SL). The Newcastle Ottawa scale
was used to grade the included documents. The case‒control study and cohort study were evaluated

through three blocks and eight items, including the selection of study population, comparability,
exposure evaluation or result evaluation. Any differences were resolved through discussion, and if
necessary, a third researcher (JX) was consulted. The maximum score is 9 stars, and studies with ≥5
stars were included in the analysis.

2.5. Statistical analysis
STATA 15.1 was used for data analysis, and P<0.05 was considered statistically significant. The
relative risk (RR) was used as the effect index for the secondary classification variable and the
combined effect quantity, and the effect quantity was expressed by the 95% confidence interval (95%

2
CI). I >50% indicated substantial heterogeneity between studies, and in such cases, the random
2
effects model was used to pool and analyze the data; when I <50%, the fixed effects model was
used to pool and analyze the data. When heterogeneity was observed, sensitivity analysis was used
to further explore the source of heterogeneity. Publication bias was evaluated by a funnel chart.

3.1. Literature retrieval process and results
A total of 637 relevant studies were retrieved. After screening, 15 articles were ultimately included,
including 13 case‒control studies, 2 cohort studies, 11 English studies and 4 Chinese studies. The

literature screening process is shown in Fig. 1.

- 64 -