Validity of Self-reported Diabetes among Middle Aged and Older Chinese Population: China Health and Retirement Longitudinal Study For peer review only

Transcription

1 Validity of Self-reported Diabetes among Middle Aged and Older Chinese Population: China Health and Retirement Longitudinal Study Journal: BMJ Open Manuscript ID: bmjopen--00 Article Type: Research Date Submitted by the Author: -Sep- Complete List of Authors: Yuan, Xiaojing; Pennington Biomedical Research Center, Chronic Disease Epidemiology Laboratory Liu, Tingting; Emory University, Nell Hodgson Woodruff School of Nursing Zou, Zhi-Yong; Peking University School of Public Health, Institute of Child and Adolescent Health Wu, Lang; Mayo Clinic, Center for Clinical and Translational Sciences Li, Changwei; Tulane University School of Public Health and Tropical Medicine, Department of Epidemiology <b>primary Subject Heading</b>: Diabetes and endocrinology Secondary Subject Heading: Research methods Keywords: DIABETES & ENDOCRINOLOGY, STATISTICS & RESEARCH METHODS, EPIDEMIOLOGY BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

2 Page of BMJ Open Validity of Self-reported Diabetes among Middle Aged and Older Chinese Population: China Health and Retirement Longitudinal Study Xiaojing Yuan *, Tingting Liu *, Zhi-Yong Zou, Lang Wu, Changwei Li. Chronic Disease Epidemiology Laboratory, Pennington Biomedical Research Center, Baton Rouge, LA, USA. Nell Hodgson Woodruff School of Nursing, Emory University, Atlanta, GA, USA. Institute of Child and Adolescent Health, School of Public Health, Peking University, Beijing, China. Center for Clinical and Translational Sciences, Mayo Clinic, Rochester, MN, USA. Department of Epidemiology, Tulane University School of Public Health and Tropical Medicine, New Orleans, LA, USA * Xiaojing Yuan and Tingting Liu contributed equally. Corresponding author: Changwei Li, MD, MPH Department of Epidemiology, Tulane University School of Public Health and Tropical Medicine, New Orleans, LA, USA Canal Street, Suite 00, New Orleans, LA 0 cli@tulane.edu Tel: Keywords: Self Reports, Diabetes Mellitus, Chinese, Sensitivity and Specificity Word count: 0 Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

3 Page of ABSTRACT Background: Self-reported diabetes has been found to be valid to evaluate people s diabetes status in the population of several countries. However, no such study has been conducted to assess the validity of self-reported diabetes in Chinese population, the largest population with the highest rate of diabetes. The aim of our study is to evaluate the validity of self-reported diabetes among middle-aged and older Chinese population. Methods: Data from 0 participants, aged, of the China Health and Retirement Longitudinal Study (CHARLS) during to were analyzed. Prevalent self-reported diabetes was compared with reference definition defined by fasting glucose, hemoglobin AC (HbAC), and medication use. Sensitivity, specificity, positive predicted value, negative predicted value, and Kappa value were calculated overall, by -year age groups, by education levels, and by living areas. Results: The sensitivity of prevalent self-reported diabetes was.%, and the specificity was.%. The sensitivity of self-reported diabetes increased with education levels, and was much higher among urban residents than rural residents (.% vs..0%). The specificity was above % among all age groups, in different education levels, and in rural and urban areas. Selfreported diabetes had substantial agreement with reference definition among participants with above vocational school education or living in urban areas (Kappa=0. and 0., respectively). Conclusions: self-reported diabetes only performed well among middle-aged and older Chinese adults with above vocational school education or among those lived in urban areas. Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

4 Page of BMJ Open STRENGTHS AND LIMITATIONS Strengths Study population was a representative sample of China s middle-aged and older population, which indicates high generalizability of the result of current study. We used fasting blood glucose and HbAc levels instead of medical records to define diabetes, which eliminated the influence of undiagnosed diabetes cases. Stringent quality control and quality assurance measures were implemented in every stage of the CHARLS study, which guaranteed the quality of our study. Limitations We included using traditional Chinese anti-diabetic medication as one of the criteria for reference definition of diabetes, which was different from the criteria which were generally accepted in studies outside China. Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

5 Page of INTRODUCTION Self-reported diagnosis of diabetes is the major method for determining diabetes in largescale epidemiologic studies [ ]. Self-reported diabetes has been found to be a valid method to evaluate people s diabetes status in the population of several countries, such as Britain[], Spain[], Brazil[], Japan[] and United States[ ]. However, no study has been conducted to examine the validity of self-reported diabetes in Chinese population, the largest population with the highest rate of diabetes [], particularly among the most vulnerable middle-aged and older Chinese population. Diabetes is one of the leading causes of mortality and morbidity worldwide, through increased risk for cardiovascular and kidney diseases [0]. In China, approximately million, or.% of adults, had diabetes in 0 []. However, only less than one-third (.%) of these patients were aware of their disease condition [], and this may affect the diabetes prevalence estimate based on self-reports in epidemiologic studies. The accuracy of self-reported information can be affected by respondents age, gender, education, medical knowledge, and frequency of contact with a physician []. The inconsistency between self-reported diabetes and the reference standard can bias the results of studies using self-reported diabetes. Typical reference definitions of diabetes include fasting glucose level, random blood glucose test, hemoglobin Ac (HbAc), oral glucose tolerance test, and/or taking diabetes medications [ ]. Therefore, the purpose of this study was to assess the validity of prevalent self-reported diabetes by using reference definition defined by blood glucose level, HbAC, and taking diabetes medication in a representative sample of middle-aged and older Chinese population. METHODS Study population Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

6 Page of BMJ Open The CHARLS surveys a representative sample of more than,000 Chinese aged from China s provinces every two years to explore the dynamics of retirement and how retirement interacts with health, health insurance, and economic well-being []. Participants were selected using a four-staged, stratified, cluster sampling method []. The baseline national survey was carried out from to through face-to-face household interview, which collected detailed information on demographics, biomedical measurements, socioeconomic status, and self-reported health status and functioning including diabetes []. The CHARLS also collected data on fasting blood glucose and HbAc. The comprehensive and multifaceted dataset provided by the CHARLS national baseline survey allow us to evaluate the validity of selfreported diabetes. The response rate among eligible households was 0.% []. Overall,, individuals aged within 0, households were investigated in the baseline survey [ ]. Blood samples were collected from, (%) out of the, study individuals by medically trained staff. A total of 0 participants had blood samples and/or information on self-reported diabetes, insulin or diabetes medication, and were included in the current analysis. The current study was secondary analysis to the publicly available data of CHARLS. The original CHARLS was approved by the Ethical Review Committee of Peking University, and all participants signed informed consent at the time of participation. Self-reported diabetes and reference definition of diabetes Participants were defined as having self-reported diabetes if they answered yes to the questions: Have you been diagnosed with diabetes or high blood sugar? The reference definition of diabetes was defined as: fasting blood glucose mg/dl, random blood glucose 0 mg/dl, HbAc.%, insulin use, or taking anti-diabetic medications including traditional Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

7 Page of Chinese medicine and modern medicine. Cut-off points for diagnosis of diabetes were based on current recommendations from the American Diabetes Association []. Measurement of blood glucose and HbAc Blood samples were collected by medically trained staff from the China Center for Disease Control and Prevention. Participants were asked to fast overnight, however, a small proportion (%) of the blood sample was also drawn even if the participants did not fast []. For those who did not fast, standards for random blood glucose level ( 0 mg/dl) and/or HbAc (.%) were used to define diabetes. After collected, plasma for glucose assay was separated from blood samples and was stored at - C [], and whole blood for HbAC assay was store immediately and during shipment at C []. All the blood samples were transported within weeks to the China Center for Disease Control and Prevention, where samples were placed at -0 C in a deep freezer before assay []. Blood assays were performed at the Youanmen Center for Clinical Laboratory of Capital Medical University during February and June []. The laboratory used quality control samples daily during the testing of the CHARLS study samples, and all test results were within the target range (within two standard deviations of mean quality control concentrations) []. Demographic variables Demographic variables including age, gender, and education were collected using standard questionnaires during the interview. Educational level was categorized as illiterate, less than primary school, primary school, middle school, high school, and above vocational school. Statistical analysis We assessed the validity of self-reported diabetes by calculating sensitivity (proportion of people with diabetes who correctly self-reported having the condition), specificity (proportion of Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

8 Page of BMJ Open non-diabetic individuals who correctly self-reported not having the condition), positive predictive value (proportion of individuals who self-reported having diabetes were diabetic based on reference definition), negative predictive value (proportion of individuals who did not report having diabetes were not diabetic based on reference definition), and Kappa value compared with the reference definition of diabetes. The validity was evaluated overall, by -year age groups, by education levels, and by rural and urban areas. Sensitivity analysis using reference definition excluding traditional Chinese anti-diabetic medication was also performed. All analyses were performed in SAS. (SAS Institute, Cary NC). RESULTS Characteristics of the participants are shown in Table. The participants had a mean age of.. Most of them had less than high school education, and.% were illiterate. The majority of the participants lived in rural areas. Overall, the participants had optimal fasting blood glucose, HbAc and body mass index. A total of.% of the participants reported having doctor diagnosed diabetes, and most of them took modern medicine to treat diabetes, however,.% of them did not treat diabetes. Based on the reference definition,.% of the participants had diabetes. Table. Characteristics of the Participants, the China Health and Retirement Longitudinal Study, -. Participants (n= 0) Age, mean (SD). (.) Male.% Education Illiterate.% Less than primary school.% Primary school.% Middle school.% High school.0% Above vocational school.% Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

9 Page of Living areas rural 0.% urban.% Self-reported diabetes.% Treatment method of diabetes Modern medicine.% Traditional Chinese medicine.% Insulin.% None.% Blood glucose, mean (SD) 0. (.) HbAC, mean (SD). (0.) Body mass index, mean (SD). (.0) SD=standard deviation; HbAC=Glycated hemoglobin The validity parameters of self-reported diabetes are presented in Table. Self-reported diabetes had high specificity (.%), positive predicted value (.%), and negative predicted value (.%). However, the sensitivity of self-reported diabetes was only.%. Overall, selfreported diabetes and the reference definition had moderate agreement (Kappa=0.0). In different -year age groups, sensitivity varied from.% to.%, specificity was all above %, positive predicted values (PPV) ranged from.% to.%, and negative predicted values (NPV) were between.% and.%. In each age group, self-reported diabetes had moderate agreement with the reference definition, and the 0 age group had the highest Kappa value (0.). The sensitivity of self-reported diabetes increased with education levels. The specificities were also all above % among participants with different education levels. PPV and NPV ranged.% -.% and.% -.%, respectively, in different education levels. Self-reported diabetes among participants with above vocational school education had substantial agreement (Kappa=0.) with the reference definition. Furthermore, participants living in rural areas had much lower sensitivity and PPV compared to those lived in urban areas, however, the specificity and NPV are similar between the two populations. Finally, urban residents had much higher Kappa value than rural residents (0. vs. 0.). Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

10 Page of BMJ Open 0 Table. Validity of Self-reported Diabetes Using Reference Definition a in the Baseline Survey of the China Health and Retirement Longitudinal Study, -. n Diabetes b, % Sensitivity Specificity PPV NPV Kappa % % CI % % CI % % CI % % CI % % CI Overall 0...,...,...,..., ,. Age groups to,...,...,...,...0,...,. 0 to,...,...0,...,. 0..,.0..,. to,...,...,...,...,.0.0.0,. 0 to,0...0,...,...,...,.0..,. to,...,...,.0.., 0.0.., 0...,. 0 to..., 0...,...,.0.., ,. to 00...,...,...,..0., 0...,. >= 0..0.,..., , ,...,. Education Illiterate,...,...0,. 0..,..., 0...,. < primary school,...,...,...,..., 0...,. Primary school,...,...,...,. 0..,...,. Middle school,0...,...,..0.,. 0.0.,...,. High School 0...,...,...,...,...,.0 > vocational school. 0..,.0..,...,..., 0...,.0 Living areas Rural,..0.,...,...,. 0.., 0...,.0 Urban,...0,...,. 0..,..0., ,. PPV=positive predicted values; NPV=negative predicted values; CI=confidence interval; a Defined as fasting glucose>= mg/dl, random glucose>=0 mg/dl, HbAC>=.%, taking oral diabetes medication (including traditional Chinese medicine and modern medicine), or using insulin; b Confirmed using the reference definition; on November by guest. Protected by copyright. Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from

11 Page 0 of 0 Table. Validity of Self-reported Diabetes Using Reference Definition a in the Baseline Survey of the China Health and Retirement Longitudinal Study, -. n Reference Diabetes, % Sensitivity Specificity PPV NPV Kappa % % CI % % CI % % CI % % CI % % CI Overall,...0,...0,...,. 0.0., 0...,. Age groups to, 0..0.,...,..0.,...,...,. 0 to,...,...,...,. 0..,.0..,. to,...,...,. 0..,...,.0..,. 0 to,0..0.,...,. 0..,...,...,. to,..0.,...,...,..., 0..0.,. 0 to...,...,...,..., 0...,. to...,.0.0.,...,..0., 0...,. >= 0...,...,...,.. 0.,...0,. Education Illiterate,...,..0.,. 0..,..., 0...,. < primary school,...,...,..0 0.,..., 0...,. Primary school,...,...,...,. 0..,...,.0 Middle school,...,...,.0..,. 0.0.,...,. High School 0...,.0..,...,.0..,...,. > vocational school...,...,...,..., 0..., 0.0 Living areas Rural,.0..,...0,...,. 0.., 0...,. Urban,...,...0,...,..., 0...0,. PPV=positive predicted values; NPV=negative predicted values; CI=confidence interval; a Defined as fasting glucose>= mg/dl, random glucose>=0 mg/dl, HbAC>=.%, taking oral diabetes medication (including only modern medicine), or using insulin; on November by guest. Protected by copyright. Page 0 of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from

12 Page of BMJ Open The sensitivity analysis using the reference definition including only modern anti-diabetic medications (Table ) showed similar specificity (.%) and NPV (0.0%). However, the sensitivity (.%) and PPV (.%) of self-reported diabetes were much smaller, and the agreement between self-reported diabetes and the reference definition was only fair (0.). DISCUSSION To the best of our knowledge, this is the first study evaluating the validity of self-reported diabetes among the middle-aged and older Chinese population. We found that self-reported diabetes was.% sensitive and.% specific. In general, self-reported diabetes did not perform well in the middle-aged and older Chinese population, however, self-reported diabetes among participants with above vocational education or living in urban areas had high sensitivity and substantial agreement with the reference definition. In the current study, self-reported diabetes largely under-estimated the prevalence of diabetes in the middle-aged and older Chinese population. This is different from studies in developed countries like the United States, where self-reported diabetes was % to 0% sensitive []. This difference are likely driven by the education level of the participants. The proportion of people who had received formal schooling higher than high school was only around 0% in the current study, which is much lower than that (above %) in the study conducted in the United States []. On contrary, sensitivities of self-reported diabetes from participants with high school or above vocational school education levels were similar to that in the United States []. Our study indicates that in large nationwide diabetes studies, blood glucose and/or HbAc tests are necessary to accurately estimate the prevalence of diabetes. Similar to studies in most countries, self-reported diabetes was highly specific in the current study [-]. In addition, the positive predicted values are very high in the current study. Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

13 Page of Hence, for studies focusing on diabetes patient only, self-report is a useful tool to recruit participants. Self-reported diabetes performed much better among urban residents compared to rural residents in the current study. This reflects the difference in education and access to medical care services between populations in the two areas. Previous studies indicated that higher education and recent health exam increased the sensitivity of self-reported diabetes []. In China, urban residents had higher education level, and better access to medical services []. Our study has several important strengths. First, the CHARLS participants were a representative sample of China s middle-aged and older population, so the generalizability of the current study is high. Second, most of the previous studies on the validity of self-reported diabetes used medical records as gold standard, which may miss undiagnosed diabetes cases, particularly for rural residents whose access to health care services is limited []. The current study use fasting blood glucose and HbAc levels to confirm diabetes, which eliminated the influence of undiagnosed diabetes cases. Third, stringent quality control and quality assurance measures were implemented in every stage of the CHARLS study. The quality of the current study can be guaranteed. Our study also has limitation. We included using traditional Chinese anti-diabetic medication in the reference definition, which was different from studies outside China. However, traditional Chinese medicine is an integral component of the health care system in China, where it is practiced side by side with conventional medicine in most hospitals and clinics []. If not considering traditional Chinese medicine in the reference definition, many cases would be missed. As in the current study, the prevalence of diabetes using the reference definition including traditional Chinese medication was very close to that in a recent large scale national survey on diabetes in 0 []. On the other hand, the diabetes prevalence using Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

14 Page of BMJ Open reference definition excluding traditional Chinese medicine was much smaller than that in the national survey in 0 []. In conclusion, self-reported diabetes among middle-aged and older Chinese population is.% specific and only.% sensitive. Self-reported diabetes had substantial agreement with the reference definition among urban residents and among participants with above vocational education. STATEMENT Changwei Li made substantial contributions to study design and data cleaning. Zhi-Yong Zou and Lang Wu helped with analytical strategy and analyzed the data. Xiaojing Yuan and Tingting Liu interpreted the results and drafted the article. All other authors revised it critically for important intellectual content. Neither of the authors have conflict of interest to disclose. There is no funding source to disclose for present study. The dataset used for current secondary analysis is publicly available on the CHARLS website ( Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

15 Page of REFERENCES. McNeely MJ, Boyko EJ. Type Diabetes Prevalence in Asian Americans Results of a national health survey. Diabetes Care 0;():-. Xaverius PK, Salas J, Kiel D. Differences in pregnancy planning between women aged -, with and without diabetes: behavioral risk factor surveillance system analysis. Diabetes research and clinical practice ;():- doi: 0.0/j.diabres..0.0[published Online First: Epub Date].. Pastorino S, Richards M, Hardy R, et al. Validation of self-reported diagnosis of diabetes in the British birth cohort. Primary care diabetes doi: 0.0/j.pcd..0.00[published Online First: Epub Date].. Espelt A, Goday A, Franch J, et al. Validity of self-reported diabetes in health interview surveys for measuring social inequalities in the prevalence of diabetes. Journal of epidemiology and community health ;():e doi: 0./jech.0.[published Online First: Epub Date].. Dode MA, Santos IS. [Validity of self-reported gestational diabetes mellitus in the immediate postpartum]. Cadernos de saude publica 0;():-. Goto A, Morita A, Goto M, et al. Validity of diabetes self-reports in the Saku diabetes study. Journal of epidemiology / Japan Epidemiological Association ;():-0. Martin LM, Leff M, Calonge N, et al. Validation of self-reported chronic conditions and health services in a managed care population. American journal of preventive medicine 00;():- Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

16 Page of BMJ Open Schneider AL, Pankow JS, Heiss G, et al. Validity and reliability of self-reported diabetes in the atherosclerosis risk in communities study. American journal of epidemiology ;():- doi: 0.0/aje/kws[published Online First: Epub Date].. Xu Y, Wang L, He J, et al. Prevalence and control of diabetes in Chinese adults. JAMA : the journal of the American Medical Association ;0():- doi: 0.00/jama..[published Online First: Epub Date]. 0. Danaei G, Finucane MM, Lu Y, et al. National, regional, and global trends in fasting plasma glucose and diabetes prevalence since 0: systematic analysis of health examination surveys and epidemiological studies with 0 country-years and. million participants. Lancet ;():- doi: 0.0/s00-()0-x[published Online First: Epub Date].. Xu Y, Wang L, He J, et al. Prevalence and control of diabetes in Chinese adults. JAMA : the journal of the American Medical Association ;0():- doi: 0.00/jama..[published Online First: Epub Date].. Okura Y, Urban LH, Mahoney DW, et al. Agreement between self-report questionnaires and medical record data was substantial for diabetes, hypertension, myocardial infarction and stroke but not for heart failure. J Clin Epidemiol 0;(0):0-0 doi: 0.0/j.jclinepi [published Online First: Epub Date].. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 0; Suppl :S-S doi: 0./dc0-S0[published Online First: Epub Date]. Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

17 Page of Zhao Y, Hu Y, Smith JP, et al. Cohort Profile: The China Health and Retirement Longitudinal Study (CHARLS). International journal of epidemiology ;():- doi: 0.0/ije/dys[published Online First: Epub Date].. Zhao Y, Strauss J, Yang G, et al. China Health and Retirement Longitudinal Study - - National Baseline Users' Guide. Beijing, China: China Center for Economic Research, Peking University,.. Zhao Y, Hi P, Hu Y, et al. China Health and Retirement Longitudinal Study, - National Baseline Blood Data Users' Guide. Beijing, China: China Center for Economic Research, Peking University,.. He W, Sengupta M, Zhang K, et al. Health and Health Care of the Older Population in Urban and Rural China: 00 International Population Reports: U.S. Census Bureau, 0:P/0-.. Hesketh T, Zhu WX. Health in China. Traditional Chinese medicine: one country, two systems. BMJ (Clinical research ed.) ;(00):- Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

18 Validity of Self-reported Diabetes among Middle-Aged and Older Chinese Adults: The China Health and Retirement Longitudinal Study Journal: BMJ Open Manuscript ID: bmjopen--00.r Article Type: Research Date Submitted by the Author: -Feb- Complete List of Authors: Yuan, Xiaojing; Pennington Biomedical Research Center, Chronic Disease Epidemiology Laboratory Liu, Tingting; Emory University, Nell Hodgson Woodruff School of Nursing Wu, Lang; Mayo Clinic, Center for Clinical and Translational Sciences Zou, Zhi-Yong; Peking University School of Public Health, Institute of Child and Adolescent Health Li, Changwei; Tulane University School of Public Health and Tropical Medicine, Department of Epidemiology <b>primary Subject Heading</b>: Diabetes and endocrinology Secondary Subject Heading: Research methods Keywords: DIABETES & ENDOCRINOLOGY, STATISTICS & RESEARCH METHODS, EPIDEMIOLOGY BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

19 Page of BMJ Open Validity of Self-reported Diabetes among Middle-Aged and Older Chinese Adults: The China Health and Retirement Longitudinal Study Xiaojing Yuan *, Tingting Liu *, Lang Wu, Zhi-Yong Zou, Changwei Li. Chronic Disease Epidemiology Laboratory, Pennington Biomedical Research Center, Baton Rouge, LA, USA. Nell Hodgson Woodruff School of Nursing, Emory University, Atlanta, GA, USA. Center for Clinical and Translational Sciences, Mayo Clinic, Rochester, MN, USA. Institute of Child and Adolescent Health, School of Public Health, Peking University, Beijing, China. Department of Epidemiology, Tulane University School of Public Health and Tropical Medicine, New Orleans, LA, USA * Xiaojing Yuan and Tingting Liu contributed equally. Co-corresponding authors Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

20 Page of Corresponding authors: Changwei Li, MD, MPH Department of Epidemiology, Tulane University School of Public Health and Tropical Medicine, New Orleans, LA, USA Canal Street, Suite 00, New Orleans, LA 0 cli@tulane.edu Tel: Zhi-Yong Zou, MD, PhD Institute of Child and Adolescent Health, School of Public Health, Peking University, Beijing, China Xueyuan Road, Beijing 000, China harveyzou0@bjmu.edu.cn Tel: +0-0 Keywords: Self Reports, Diabetes Mellitus, Chinese, Sensitivity and Specificity Word count: 0 Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

21 Page of BMJ Open Abstract Background: Self-reported diabetes has been found to be valid to evaluate people s diabetes status in the population of several countries. However, no such study has been conducted to assess the validity of self-reported diabetes in Chinese population, the largest population with the highest rate of diabetes. The aim of our study is to evaluate the validity of self-reported diabetes among middle-aged and older Chinese population. Methods: Data from 0 participants, aged, of the China Health and Retirement Longitudinal Study (CHARLS) during to were analyzed. Prevalent self-reported diabetes was compared with reference definition defined by fasting glucose, hemoglobin AC (HbAC), and medication use. Sensitivity, specificity, positive predicted value, negative predicted value, and Kappa value were calculated overall, by -year age groups, by education levels, and by living areas. Results: The sensitivity of prevalent self-reported diabetes was.%, and the specificity was.%. The sensitivity of self-reported diabetes increased with education levels, and was much higher among urban residents than rural residents (.% vs..0%). The specificity was above % among all age groups, in different education levels, and in rural and urban areas. Selfreported diabetes had substantial agreement with reference definition among participants with above vocational school education or living in urban areas (Kappa=0. and 0., respectively). Conclusions: Although the sensitivity of self-reported diabetes was poor among middleaged and older Chinese adults, the specificity and positive predictive values were fairly good. Furthermore, self-reported diabetes performed well among those with more than vocational school educations or living in urban areas. Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

22 Page of Strengths and Limitations Strengths Study population was a representative sample of China s middle-aged and older population, which indicates high generalizability of the result of current study. We used fasting blood glucose and HbAc levels instead of medical records to define diabetes, which eliminated undiagnosed cases of diabetes. Stringent quality control and quality assurance measures were implemented in every stage of the CHARLS study, which guaranteed the quality of our study. Limitations We included using traditional Chinese anti-diabetic medication as one of the criteria for reference definitions of diabetes, which was different from the criteria which were generally accepted in studies outside China. Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

23 Page of BMJ Open Introduction Self-reported diagnosis of diabetes is the major method for determining diabetes in largescale epidemiologic studies [ ]. Self-reported diabetes has been found to be a valid method to evaluate people s diabetes status in the population of several countries, such as Britain[], Spain[], Brazil[], Japan[] and United States[ ]. However, no study has been conducted to examine the validity of self-reported diabetes in China, which has the largest number of people with diabetes in the world [], particularly among the most vulnerable middle-aged and older Chinese population. Diabetes is one of the leading causes of mortality and morbidity worldwide, which increased the risk for cardiovascular and kidney diseases [0]. In China, approximately million, or.% of adults, had diabetes in 0 []. However, less than one-third (.%) of these patients were aware of their disease condition [], and this may affect the estimate of diabetes prevalence based on self-reports in large-scale epidemiologic studies. The accuracy of self-reported information can be affected by respondents age, gender, education, medical knowledge, and frequency of contact with a physician []. The inconsistency between selfreported diabetes and the reference standard can bias the results of studies using self-reported diabetes. Typical reference definitions of diabetes include fasting glucose level, random blood glucose test, hemoglobin Ac (HbAc), oral glucose tolerance test, and/or taking diabetes medications [ ]. Therefore, the purpose of this study was to assess the validity of prevalent self-reported diabetes by using reference definitions defined by fasting or random blood glucose levels, HbAc, and taking diabetes medication in a representative sample of middle-aged and older Chinese population using baseline data from the China Health and Retirement Longitudinal Study (CHARLS). Methods Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

24 Page of Subjects and Design The CHARLS surveys a representative sample of more than,000 Chinese aged from China s provinces every two years to explore the dynamics of retirement and how retirement interacts with health, health insurance, and economic well-being []. Participants were selected using a four-staged, stratified, cluster sampling method []. The baseline national survey was carried out from to through face-to-face household interview, which collected detailed information on demographics, biomedical measurements, socioeconomic status, and self-reported health status and functioning including diabetes []. The CHARLS also collected data on fasting blood glucose and HbAc. The comprehensive and multifaceted dataset provided by the CHARLS national baseline survey allow us to evaluate the validity of selfreported diabetes. The response rate among eligible households was 0.% []. Overall,, individuals aged within 0, households were investigated in the baseline survey [ ]. Blood samples were collected from, (%) out of the, study individuals by medically trained staff. A total of 0 participants had blood samples and/or information on self-reported diabetes, insulin or diabetes medication, and were included in the current analysis. The current study was secondary analysis of the open-access dataset of the CHARLS. The original CHARLS was approved by the Ethical Review Committee of Peking University, and all participants signed informed consent at the time of participation. Self-reported diabetes and reference definitions of diabetes Participants were defined as having self-reported diabetes if they answered yes to the questions: Have you been diagnosed with diabetes or high blood sugar? The reference definitions of diabetes were defined as: fasting blood glucose mg/dl, or random blood glucose 0 mg/dl, or HbAc.%, or insulin use, or taking anti-diabetic medications Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

25 Page of BMJ Open including traditional Chinese medicine and modern medicine. The cut-off points for diagnosis of diabetes were based on current recommendations from the American Diabetes Association []. Measurement of blood glucose and HbAc Blood samples were collected by medically trained staff from the China Center for Disease Control and Prevention. Participants were asked to fast overnight. However, a small proportion (%) of the blood sample was also drawn even if the participants did not fast []. For those who did not fast, standards for random blood glucose level ( 0 mg/dl) and/or HbAc (.%) were used to define diabetes. After collected, plasma for glucose assay was separated from blood samples and was stored at - C [], and whole blood for HbAC assay was store immediately and during shipment at C []. All the blood samples were transported within weeks to the China Center for Disease Control and Prevention, where samples were placed at -0 C in a deep freezer before assay []. Blood assays were performed at the Youanmen Center for Clinical Laboratory of Capital Medical University during February and June []. The laboratory used quality control samples daily during the testing of the CHARLS study samples, and all test results were within the target range (within two standard deviations of mean quality control concentrations). Glucose was measured using enzymatic colormetric test, and HbAc was analyzed using Boronate affinity chromatography []. Demographic variables Demographic variables including age, gender, and educational level were collected using standardized questionnaires during the interview. Educational level was categorized as illiterate, less than primary school, primary school, middle school, high school, and above vocational school. Statistical Analysis Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

26 Page of We assessed the validity of self-reported diabetes by calculating sensitivity (proportion of people with diabetes who correctly self-reported having the condition), specificity (proportion of individuals without diabetes who correctly self-reported not having the condition), positive predictive value (proportion of individuals who self-reported having diabetes were diabetic based on reference definitions), negative predictive value (proportion of individuals who did not report having diabetes were not diabetic based on reference definitions), and Kappa value compared with the reference definitions of diabetes. The validity was evaluated overall, by -year age groups, by education levels, and by rural and urban areas. Sensitivity analysis using reference definitions excluding traditional Chinese anti-diabetic medication was also performed. All analyses were performed in SAS. (SAS Institute, Cary NC). Results Characteristics of the participants are shown in Table. The participants had a mean age of. years old. Most of them had less than high school education, and.% were illiterate. The majority of the participants (0.%) lived in rural areas. Overall, the participants had optimal fasting blood glucose, HbAc and body mass index. A total of.% of the participants reported physician-diagnosed diabetes, and most of them took modern medicine to treat diabetes, however,.% of them did not treat diabetes. Based on the reference definitions,.% of the participants had diabetes. Table. Baseline Characteristics of the Participants, the China Health and Retirement Longitudinal Study, -. Participants (n= 0) Age, mean (SD). (.) Male.% Education Illiterate.% Less than primary school.% Primary school.% Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

27 Page of BMJ Open Middle school.% High school.0% Above vocational school.% Living areas rural 0.% urban.% Self-reported diabetes.% Confirmed diabetes using definition *, %.% Confirmed diabetes using definition, %.% Treatment method of diabetes Modern medicine.% Traditional Chinese medicine.% Insulin.% None.% Blood glucose. mg/dl, mean (SD) 0. (.) HbAc, %, mean (SD). (0.) Body mass index, kg/m, mean (SD). (.0) SD=standard deviation * defined as fasting glucose>= mg/dl or random glucose>=0 mg/dl or HbAC>=.% or taking oral diabetes medication (including only modern medicine) or using insulin; defined as fasting glucose>= mg/dl or random glucose>=0 mg/dl or HbAC>=.%, taking oral diabetes medication (including traditional Chinese medicine and modern medicine) or using insulin; The validity parameters of self-reported diabetes are presented in Table. Self-reported diabetes had high specificity (.%), positive predicted value (.%), and negative predicted value (.%). However, the sensitivity of self-reported diabetes was only.%. Overall, selfreported diabetes and the reference definitions had moderate agreement (Kappa=0.0). In different -year age groups, sensitivity varied from.% to.%, specificity was all above %, positive predicted values (PPV) ranged from.% to.%, and negative predicted values (NPV) were between.% and.%. In each age group, self-reported diabetes had moderate agreement with the reference definitions, with the 0- age group having the highest Kappa value (0.). The sensitivity of self-reported diabetes increased with education levels. The specificities were also all above % among participants with different education levels. PPV and Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

28 Page 0 of NPV ranged.-.% and.-.%, respectively, in different education levels. Selfreported diabetes among participants with above vocational school education had substantial agreement (Kappa=0.) with the reference definition. Furthermore, participants living in rural areas had much lower sensitivity and PPV compared to those living in urban areas. However, the specificity and NPV are similar between the two populations. Finally, urban residents had much higher Kappa value than rural residents (0. vs. 0.). The sensitivity analysis using the reference definitions including only modern antidiabetic medications (Table ) showed similar specificity (.%) and NPV (0.0%). However, the sensitivity (.%) and PPV (.%) of self-reported diabetes were much smaller, and the agreement between self-reported diabetes and the reference definitions was only fair (0.). Page 0 of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

29 Page of BMJ Open 0 Table. Validity of Self-reported Diabetes Using Reference Definitions a in the Baseline Survey of the China Health and Retirement Longitudinal Study, -. n Diabetes b, % Sensitivity Specificity PPV NPV Kappa % % CI % % CI % % CI % % CI % % CI Overall 0...,...,...,..., ,. Age groups to,...,...,...,...0,...,. 0 to,...,...0,...,. 0..,.0..,. to,...,...,...,...,.0.0.0,. 0 to,0...0,...,...,...,.0..,. to,...,...,.0.., 0.0.., 0...,. 0 to..., 0...,...,.0.., ,. to 00...,...,...,..0., 0...,. >= 0..0.,..., , ,...,. Education Illiterate,...,...0,. 0..,..., 0...,. < primary school,...,...,...,..., 0...,. Primary school,...,...,...,. 0..,...,. Middle school,0...,...,..0.,. 0.0.,...,. High School 0...,...,...,...,...,.0 > vocational school. 0..,.0..,...,..., 0...,.0 Living areas Rural,..0.,...,...,. 0.., 0...,.0 Urban,...0,...,. 0..,..0., ,. PPV=positive predicted values; NPV=negative predicted values; CI=confidence interval; a Defined as fasting glucose>= mg/dl or random glucose>=0 mg/dl or HbAC>=.% or taking oral diabetes medication (including traditional Chinese medicine and modern medicine) or using insulin; b Confirmed using the reference definitions; on November by guest. Protected by copyright. Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from

30 Page of 0 Table. Validity of self-reported diabetes in the baseline survey of CHARLS using the reference definition not including taking Traditional Chinese medicine a n Reference Diabetes, % Sensitivity Specificity PPV NPV Kappa % % CI % % CI % % CI % % CI % % CI Overall,...0,...0,...,. 0.0., 0...,. Age groups to, 0..0.,...,..0.,...,...,. 0 to,...,...,...,. 0..,.0..,. to,...,...,. 0..,...,.0..,. 0 to,0..0.,...,. 0..,...,...,. to,..0.,...,...,..., 0..0.,. 0 to...,...,...,..., 0...,. to...,.0.0.,...,..0., 0...,. >= 0...,...,...,.. 0.,...0,. Education Illiterate,...,..0.,. 0..,..., 0...,. < primary school,...,...,..0 0.,..., 0...,. Primary school,...,...,...,. 0..,...,.0 Middle school,...,...,.0..,. 0.0.,...,. High School 0...,.0..,...,.0..,...,. > vocational school...,...,...,..., 0..., 0.0 Living areas Rural,.0..,...0,...,. 0.., 0...,. Urban,...,...0,...,..., 0...0,. PPV=positive predicted values; NPV=negative predicted values; CI=confidence interval; a Defined as fasting glucose>= mg/dl or random glucose>=0 mg/dl or HbAC>=.% or taking oral diabetes medication (including only modern medicine) or using insulin; on November by guest. Protected by copyright. Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from

31 Page of BMJ Open Discussion To the best of our knowledge, this is the first study evaluating the validity of self-reported diabetes among the middle-aged and older Chinese population. We found that self-reported diabetes was.% sensitive and.% specific. In general, self-reported diabetes did not perform well in the middle-aged and older Chinese population. However, self-reported diabetes among participants with above vocational education or living in urban areas had high sensitivity and substantial agreement with the reference definitions. In the current study, self-reported diabetes largely under-estimated the prevalence of diabetes in the middle-aged and older Chinese population. This is different from studies reported in developed countries such as the United States, where self-reported diabetes was % to 0% sensitive []. This difference is likely driven by the education level of the participants. The proportion of people who had received formal schooling higher than high school was only around 0% in the current study, which is much lower than that (above %) in the study conducted in the United States []. On contrary, sensitivity of self-reported diabetes from participants with high school or above vocational school education levels was similar to that in the United States []. Our study indicates that blood glucose and/or HbAc tests are necessary to accurately estimate the prevalence of diabetes in large nationwide diabetes studies. Similar to studies in most countries, self-reported diabetes was highly specific in the current study [-]. In addition, the positive predicted values are very high in the current study. Hence, for studies with patients with diabetes, self-report is a useful tool to identify and recruit participants. Self-reported diabetes performed much better among urban residents compared to rural residents in the current study. This may reflect differences in education and access to health care services between populations in these areas. Previous studies indicated that higher education and Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.

32 Page of recent health exam increased the sensitivity of self-reported diabetes []. In China, urban residents had higher education level, and better access to medical services []. Our study has several important strengths. First, the CHARLS participants were a representative sample of China s middle-aged and older population. So the findings can be generalized to all middle-aged and older Chinese adults. Second, most of the previous studies on the validity of self-reported diabetes used medical records as gold standard, which may miss cases of undiagnosed diabetes, particularly for rural residents whose access to health care services is limited []. The current study use fasting blood glucose and HbAc levels to confirm diabetes, which eliminated cases of undiagnosed diabetes. Third, stringent quality control and quality assurance measures were implemented in every stage of the CHARLS study. Therefore, the quality of the current study can be guaranteed. Our study also has limitations. First, the response rate of blood sample collection was a little bit low (%), and according to the guideline of CHARLS, younger men were more likely to be missed in blood sample collection, mostly due to time conflict with their work. Since younger people have lower diabetes prevalence, lower response rate of this population will decrease the negative predictive values. Second, we included using traditional Chinese anti-diabetic medication in the reference definitions, which was different from studies outside China. However, traditional Chinese medicine is an integral component of the health care system in China, where it is practiced side by side with conventional medicine in most hospitals and clinics []. Many patients seek both modern medicine and traditional Chinese medicine to manage diabetes. If not considering traditional Chinese medicine in the reference definitions, many cases would be missed. As in the current study, the prevalence of diabetes using the reference definitions including traditional Chinese medication was very close to that in a recent large scale national survey on diabetes in 0 []. On the other hand, the diabetes prevalence using reference definitions excluding traditional Page of BMJ Open: first published as 0./bmjopen--00 on April. Downloaded from on November by guest. Protected by copyright.