随机研究偏倚评估数据集

数据集获取（Figshare）

• 源站标题：Bias assessment for randomized studies.
• DOI：10.1371/journal.pone.0291931.t005
• 发布日期：2024-02-21
• 文件数：1 个
• 文件总体积：5.63 KB
• 源站页面：https://plos.figshare.com/articles/dataset/Bias_assessment_for_randomized_studies_/25261412

命令行下载

curl -L -o "Table 5.xls" "https://ndownloader.figshare.com/files/44627147"

文件清单

• Table 5.xls（5.63 KB）

文件总体积 5.63 KB（源站 API 实测）。

数据加载示例（图像类）

from PIL import Image
import glob, os

files = (glob.glob(os.path.join(path, "**", "*.png"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.jpg"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.tif"), recursive=True))
print("图像文件数:", len(files))
img = Image.open(files[0]); print("尺寸/模式:", img.size, img.mode)

# torchvision Dataset 方式:
# from torchvision import datasets
# ds = datasets.ImageFolder(path)  # 要求 子目录=类别

目录组织与标注格式以源站说明和下载后实际文件为准。

数据说明

Introduction
Ensuring timely follow-up of abnormal screening results is essential for eliminating cervical cancer.
Objective
The purpose of the study was to review single and multicomponent interventions designed to improve follow-up of women with abnormal cervical cancer screening results. We report on effectiveness across studies, and describe what aspects of these interventions might be more impactful.
Methods
Publications were searched between January 2000 and December 2022. The search included observational, quasi-experimental (pre-post studies) and randomized controlled studies describing at least one intervention to increase follow-up of women with abnormal cervical cancer screening results. Outcomes of studies included completion of any follow-up (i.e., attending a follow-up appointment), timely diagnosis (i.e., colposcopy results within 90 days of screening) and time to diagnostic resolution (i.e., days between screening and final diagnosis). We assessed risk of bias for observational and quasi-experimental studies using the Newcastle-Ottawa Scale (NOS) tool and the Cochrane collaboration tool for randomized studies. We conducted a meta-analysis using studies where data were provided to estimate a summary average effect of the interventions on follow-up of patients and to identify characteristics of studies associated with an increased effectiveness of interventions. We extracted the comparison and intervention proportions of women with follow-up before and after the intervention (control and intervention) and plotted the odds ratios (ORs) of completing follow-up along with the 95% confidence intervals (CIs) using forest plots for the interventions vs. controls when data were available.
Findings
From 7,457 identified studies, 28 met the inclusion criteria. Eleven (39%) of the included studies had used a randomized design. Most studies (63%) assessed completion of any follow-up visit as the primary outcome, whereas others measured time to definite diagnosis (15%) or diagnostic resolution (22%). Navigation was used as a type of intervention in 63% of the included studies. Most interventions utilized behavioral approaches to improve outcomes. The overall e

许可

CC BY 4.0

引用

学术使用请引用 DOI 10.1371/journal.pone.0291931.t005（source: https://plos.figshare.com/articles/dataset/Bias_assessment_for_randomized_studies_/25261412）。