子宮頸癌及乳癌篩檢服務利用之城鄉差距探討

Abstract

背景：癌症（惡性腫瘤）為我國十大死因之首，自1982年起至2022年已連續41年居國人死因之首，且癌症死亡人數及死亡率皆有逐年上升的趨勢。然而實證研究證實癌症篩檢可以有效降低癌症死亡率，規律的預防照護可降低死亡率、醫療服務利用及醫療成本。台灣目前針對女性提供的癌症篩檢補助服務包括子宮頸抹片檢查及乳房X光攝影檢查，雖為公費補助之篩檢項目，但是台灣有醫療資源分布不均之現象，癌症篩檢服務是否能觸及所有符合篩檢條件的女性乃是一值得探討的問題。台灣仍缺乏針對城鄉婦女子宮頸癌及乳癌篩檢服務利用率差異探討之相關研究，城鄉篩檢情況是否存在差異仍為未知，因此本研究欲針對此面向進行分析及探討。

目的：本研究針對台灣居住於城市及偏鄉婦女的乳癌及子宮頸癌篩檢情況進行統計分析，探討偏鄉婦女與城市婦女之乳癌篩檢情況與子宮頸癌篩檢情況是否存在差異。

方法：本研究採用回溯性世代研究法(Retrospective Cohort study)，利用全民健保資料庫進行次級資料分析。研究對象為2017年年滿30歲及以上女性（子宮頸癌篩檢）及2017年45-69歲女性（乳癌篩檢）。利用衛生福利部衛生福利資料科學中心全民健康保險資料庫2017年至2019年三年全人口資料檔，包含「Health01_全民健保處方及治療明細檔」、「Health07_全民健保承保檔」、「Health10_死因統計檔」及「Health45_癌症登記年報檔」進行分析。利用承保檔中投保單位地區、保險單位類別、被保人身份、呼吸道感染門診就醫紀錄等推估居住地，之後再進行居住地城鄉劃分。並針對研究對象居住地（城/鄉）與篩檢情況進行雙變項卡方檢定及多變項羅吉斯迴歸分析，探討居住地不同是否造成癌症篩檢利用情況差異。

結果：本研究之雙變量結果發現，2017-2019年整體子宮頸癌篩檢率為48.02%，城市女性子宮頸癌篩檢率為48.00%，而偏鄉女性子宮頸癌篩檢率為48.16%，高於整體子宮頸癌篩檢率及城市女性子宮頸癌篩檢率，且皆達統計顯著差異（p<0.0001）。研究對象之年齡、投保金額、地區教育程度與是否進行子宮頸癌篩檢亦呈顯著相關（p<0.0001）。近一步進行多變量羅吉斯迴歸分析，在控制年齡、投保金額、地區教育程度、共病症情形後，以居住地於城市者作為對照組，居住於偏鄉進行子宮頸癌篩檢之勝算比為1.017（95%CI [1.011,1.024]），達統計顯著差異（p<0.0001）。此外，研究結果亦發現隨著年齡增加接受子宮頸癌篩檢之勝算比越低，投保金額較高者更可能進行子宮頸癌篩檢（OR>1），教育程度較高者卻更不可能進行子宮頸癌篩檢（OR<1）。乳癌篩檢部分，2017-2018年整體乳癌篩檢率為38.86%，城市女性乳癌篩檢率為38.80%，而偏鄉女性乳癌篩檢率為39.56%，高於整體乳癌篩檢率及城市女性乳癌篩檢率，且皆達統計顯著差異（p<0.0001）。研究對象之年齡、投保金額、教育程度與是否進行乳癌篩檢亦呈顯著相關（p<0.0001）。近一步進行多變量羅吉斯迴歸分析，在控制年齡、投保金額、地區教育程度、共病症情形後，以居住地於城市者作為對照組，居住於偏鄉進行乳癌篩檢之勝算比為1.038（95%CI [1.029,1.048]），達統計顯著差異（p<0.0001）。此外，年齡、投保金額、教育程度較高者均更可能接受乳癌篩檢（OR>1）。

結論：本研究發現2017至2019年，台灣婦女在子宮頸癌及乳癌篩檢利用上確實存在城鄉差距，但是偏鄉之篩檢狀況並不比城市差，反而出現城市篩檢狀況不及偏鄉的狀況。年齡、投保金額、地區教育程度及共病症情形也均為可預測研究對象篩檢行為之相關因素。本研究亦發現子宮頸癌及乳癌的篩檢率整體偏低，子宮頸癌篩檢率為48.02%，乳癌篩檢率為38.86%，與國家癌症防治計畫目標之間仍有差距，應設法國家提高整體篩檢率。

Background: Cancer (malignant tumors) has been the leading cause of death in Taiwan, holding this position for 41 consecutive years from 1982 to 2022. Both the number of cancer deaths and the mortality rate have shown an upward trend over the years. However, empirical studies have confirmed that cancer screening can effectively reduce cancer mortality. Regular preventive care can lower mortality rates, medical service utilization, and healthcare costs. Currently, Taiwan provides subsidized cancer screening services for women, including Pap smears and Mammograms. Although these screenings are publicly funded, the uneven distribution of medical resources in Taiwan raises concerns about whether cancer screening services are accessible to all eligible women. Taiwan still lacks relevant research on the differences in the utilization rates of cervical and breast cancer screening services between urban and rural women. Whether there are disparities in screening rates between urban and rural areas remains unknown. Therefore, this study aims to analyze and explore this aspect.

Objectives: This study conducts a statistical analysis of breast cancer screening and cervical cancer screening among women residing in urban and rural areas in Taiwan. We aim to explore whether there are differences in breast and cervical cancer screening between rural and urban women.

Methods: This study employs a Retrospective Cohort study design and utilizes the National Health Insurance Research Database (NHIRD) for secondary data analysis. The study subjects are women aged 30 and above for cervical cancer screening and women aged 45-69 for breast cancer screening in 2017. Data from the Health and Welfare Data Science Center (HWDC), covering the entire population from 2017 to 2019, is used. This includes the "Health01_Ambulatory Care Expenditures by Visits," "Health07_ Registry for Beneficiaries," "Health10_Cause of Death Data," and "Health45_Cancer Registry Annual Report Files" for analysis. The Registry for Beneficiaries file is used to estimate the place of residence based on the insurance unit's region, insurance unit type, insured person's identity, and outpatient records of respiratory infections. The place of residence is then classified into urban and rural areas. Chi-square tests and multivariate logistic regression analyses are conducted to examine the differences in cancer screening utilization based on the place of residence (urban/rural).

Results: The bivariate results of this study found that the overall cervical cancer screening rate from 2017 to 2019 was 48.02%. The cervical cancer screening rate for urban women was 48.00%, while for rural women it was 48.16%, higher than the overall cervical cancer screening rate and the rate for urban women, with all differences being statistically significant (p<0.0001). Age, insurance amount, and regional education level of the study subjects were also significantly associated with cervical cancer screening (p<0.0001). Further multivariate logistic regression analysis, controlling for age, insurance amount, regional education level, and comorbidities, showed that compared to those living in urban areas, the odds ratio for cervical cancer screening in rural areas was 1.017 (95% CI [1.011, 1.024]), which was statistically significant (p<0.0001). Furthermore, the study results also found that the odds ratio for cervical cancer screening decreases with increasing age, those with higher insurance amount are more likely to undergo cervical cancer screening (OR>1), but those with higher regional education levels are less likely to undergo cervical cancer screening (OR<1). For breast cancer screening, the overall screening rate from 2017 to 2018 was 38.86%. The screening rate for urban women was 38.80%, while for rural women it was 39.56%, higher than the overall breast cancer screening rate and the rate for urban women, with all differences being statistically significant (p<0.0001). Age, insurance amount, and regional education level of the study subjects were also significantly associated with breast cancer screening (p<0.0001). Further multivariate logistic regression analysis, controlling for age, insurance amount, regional education level, and comorbidities, showed that compared to those living in urban areas, the odds ratio for breast cancer screening in rural areas was 1.038 (95% CI [1.029, 1.048]), which was statistically significant (p<0.0001). Furthermore, individuals who are older, have higher insurance amount, and higher regional education levels are more likely to undergo breast cancer screening (OR>1).

Conclusions: This study found that from 2017 to 2019, there was a disparity in cervical and breast cancer screening utilization between urban and rural areas among Taiwanese women. However, the screening rates in rural areas were not lower than those in urban areas; in fact, urban areas exhibited lower screening rates compared to rural areas. Factors such as age, insurance amount, regional education level, and comorbidity status were all related to predicting screening behaviors among the study subjects. The study also revealed that the overall screening rates for cervical and breast cancer were low, with cervical cancer screening at 48.02% and breast cancer screening at 38.86%. These rates fall short of the targets set by the National Cancer Control Programme - Cancer Screening, highlighting the need to improve overall screening rates.