臺灣醫師與牙醫師死亡風險之研究

Mortality Rates and their Determinants among Physicians and Dentists in Taiwan

商東福(Tung-Fu Shang)

指導教授 ： 王榮德

國立臺灣大學/公共衛生學院/職業醫學與工業衛生研究所/博士(2011年)

https://doi.org/10.6342/NTU.2011.00189

摘要

前言：雖然有某些專科別醫師被分析出是自殺、藥物濫用及癌症的高風險職業，國外學者的研究大多發現醫師的死亡率比一般老百姓來的低。不過醫師的工作時間長、壓力也大，某些專科別的醫師甚至有長期暴露在感染性血液，體液，空氣或低量放射線的職業風險之中，所以，臺灣醫師的死亡風險是否比一般人民高一直都是民眾關切的事情。2003年SARS疫情也讓人想到各專科別醫師接觸科別常見癌症及傳染性疾病，可能透過病毒、細菌傳遞致病，例如治療肺癌專家罹患肺癌，肝膽專家死於肝癌等等。因此，本研究應用流行病學的方法並以社經地位相當的內科醫師做為對照，來分析臺灣醫師、牙醫師及不同專科醫師的死亡風險。
研究方法與對象：本研究對象取自中華民國醫師及牙醫師公會全國聯合會會員檔，包括民國79年到95年間曾經在臺灣執業的37,545位醫師，及民國74年到98年的11,700位牙醫師。本研究使用身份證字號和衛生署建立的死亡原因檔，進行比對及記錄死亡原因；以民國60年到95年間的台灣人及同一資料庫的18,664位內科醫師為對照族群，使用生命表分析系統計算不同死因的標準化死亡比；並運用Cox 比例風險回歸模式，考慮執業的時間，校正包括執業地區、開始執業時年齡及年代等因素，評估不同專科別與死亡的相關性。
結果：校正分析性別、執業年齡、年代等因素後，本研究發現臺灣醫師包括各專科別醫師、牙醫師各種疾病的標準化死亡比比一般老百姓低，自殺與藥物濫用的標準化死亡比遠低於1，分別為0.14與0.16。當我們比較不同專科醫師之間的死亡風險差異時，發現麻醉科、外科醫師，比內科醫師高，麻醉醫師又高於外科醫師，死亡風險分別為1.97與1.23。當用內科醫師做為對照族群時牙醫師各種疾病的標準化死亡比略高於1 (1.13) ，心臟病及溺水的標準化死亡比也高分別為1.66和6.62。進一步控制相關影響因素之後牙醫師的死亡風險高於內科醫師為1.17。另外以醫師人口比1:500為對照，醫師人口比愈低，該地區執業醫師的死亡風險也愈高。
結論：臺灣醫師、牙醫師整體疾病的死亡風險都明顯低於一般老百性，而且也沒有存在國外研究所發現的高自殺死亡風險。本研究可以澄清臺灣坊間長期以來關於醫師短命的傳言與印象。若以內科醫師做為對照，牙醫師以及外科、麻醉科醫師的死亡風險高，進一步針對死亡原因的分析，除了牙醫師心臟疾病及溺水意外的標準化死亡比較高，其餘可能與職場暴露有關的癌症和感染的標準化死亡比並未發現增加的情形。醫師執業地區的地域差異和醫師人口比除了影響該地區居民的健康，也對醫師本身的死亡風險產生同步影響，增加執業醫師人力有助於地區民眾與醫師本身的健康。

關鍵字

醫師 ； 牙醫師 ； 死亡 ； 不均 ； 期望壽命 ； 標準化死亡比 ； 生命表分析系統 ； 醫師人口比 ； 死亡決定因子 ； Cox 比例風險回歸

並列摘要

Background and Purpose: There have been many studies in industrialized countries that demonstrate that medical professionals experience lower overall mortality rates than other occupations. However, there are also controversial claims on the potentially higher mortality rates among physicians than the general public due to long working hours in high stress environments with frequent exposure to physical and biological agents. During 2003 SARS epidemic further linked some of the physicians are at the risks of infection or cancer due to physical contacts of the related diseases. For example, the Pulmonologists, especially experts for lung cancer suffered from lung malignancy. And liver cancer specialists died from hepatoma. Several studies have described larger health problems and higher death rates in certain specialties such as surgery, with causes being attributed to suicide, drug abuse, cancer, etc. The aim of the study is to analyze the survival data of all Taiwanese physicians by department, clarify the suspicion of premature mortality among physicians exposed to different agents, and further identify predictors of mortality among them.
Methods: Data from 37,545 physicians registered in the database of Taiwan Medical Association (TMA) during 1990 to 2006 and data from 11,700 dentists registered in the Taiwan Dental Association (TDA) during 1985 to 2009 were enrolled for analysis. Overall and cause-specific standardized mortality ratios (SMRs) of these individuals were obtained from the National Mortality Database using the IDs of the cases. Using the Life Table Analysis System (LTAS), the mortality rates and associated figures of these cases were compared with first the national death rates of all Taiwan citizens in 1971 to 2006 and then a set of 18,664 internists. A multivariate mortality rate analysis was also performed by the Cox’s proportional hazards regression, using survival and years of practice to calculate relationships to specialties, gender, geographic region of practices, regional health resources, ages of initial practices, and years of initial practice
Results: Compared with all Taiwanese citizens, physicians of all medical specialties and departments in Taiwan were found to be less likely to die from all causes regardless of age, gender, or years of occupation. In particular, the SMRs for suicide and drug abuse were generally below 0.50, with suicide’s SMR at 0.14 (95% Confidence Interval =0.09 - 0.21) and drug abuse’s SMR at 0.16 (95% CI= 0.07 - 0.32).
The Cox regression model showed that the anesthesiologists had the highest hazard ratio (HR) of 1.97, seconded by surgeons at 1.23. However, the overall SMRs for surgeons and anesthesiologists were only marginally elevated at 1.15 (95% CI=0.98-1.34) and 1.62 (95% CI=0.93-2.64) respectively.
When internists were chosen as the reference group, significant excess mortalities (greater than 1) were observed in dentists, with the overall cause SMR at 1.13 (95% CI = 1.00-1.26), drowning SMR at 6.62 (95% CI=2.15-15.45), and heart disease SMR at 1.66 (95% CI = 1.22-2.21). The Cox regression model showed that the dentists experienced a higher HR of 1.17 (95% CI, 1.01 to 1.37) when compared to internists as well. In addition, compared with a physician to population ratio of 1:500, it was discovered that the lower the number of physicians in a population, the higher the HR of physicians are.
Conclusions: The risks of cause-specific mortality of physicians with different medical specialties were found to be significantly lower than those of the general population in Taiwan, which serves to rectify the perception that physicians are subject to premature mortality. Additionally, the study found that physicians in Taiwan do not have the significantly raised suicide HR as previously published overseas. Compared with internists, Taiwanese dentists had significantly elevated SMRs for overall causes, drowning, and heart disease. Careful precaution should be taken to reduce the trend. Future studies and analysis should be performed to explore the mechanisms of how professional stress and exposures contribute to the increased mortality risks in Taiwanese dentists. The physician to population ratio and the geographic region of physician's practice may result in disparities of physician mortality. Increasing the numbers of physicians and/or improving the practice environment may help to reduce the health disparities of both the general public and physicians residing in a region with poor resources.

並列關鍵字

Physician ； dentist ； disparity ； mortality rate ； life expectancy ； standardized mortality ratios (SMRs) ； life Table Analysis System (LTAS) ； the physician to population ratio ； determinants of mortality ； Cox’s proportional hazards regression

參考文獻

21. Shih YT, Hung YT, Chang HY, et al. The design, contents, operation and the characteristics of the respondents of the 2001 National Health Interview Survey in Taiwan. Taiwan J Public Health 2003;22:465-73.

19. Lin YW, Chang YW, Tsai CC. Job Strain and Health-Related Quality of Life of Hospital Employees: Case of a Medical Center in Taichung. Taiwan Journal of Public Health 2004;23 (2):108-20.

13. National Health Insurance. Statistical Annual Report of Medical Care. Taipei: The ROC Department of Health; 2005.

14. Division of Health Statistics. Trend of Health Statistics in Taiwan. Taipei: The ROC Department of Health; 2005.

18. National Health Insurance. Statistical Annual Report of Medical Care. Taipei: The ROC Department of Health; 2003.