BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from
BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available.
When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to.
The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript.
BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay-per-view fees (http://bmjopen.bmj.com).
If you have any questions on BMJ Open’s open peer review process please email [email protected] http://bmjopen.bmj.com/ on September 23, 2021 by guest. Protected copyright. BMJ Open BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from
Health Literacy and Quality of Life Among Cancer Survivors in China
Journal: BMJ Open ManuscriptFor ID peerbmjopen-2018-028458 review only Article Type: Research
Date Submitted by the 11-Dec-2018 Author:
Complete List of Authors: Xia, Juan; Fudan University, school of public health Deng, Qinglong; Fudan University, school of public health Yan, Rui; Fudan University, school of public health Wu, Peng; Fudan University, school of public health Yang, Renren; Fudan University, school of public health Lv, Binghui; Fudan University, school of public health Wang, Jiwei; Fudan University, Fudan university school of public health Yu, Jinming; Fudan University, school of public health
Keywords: Health Literacy, 3 Brief Questions, Quality of Life, Cancer Survivors
http://bmjopen.bmj.com/ on September 23, 2021 by guest. Protected copyright.
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 18 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Title Page 5 6 7 Title: Health Literacy and Quality of Life Among Cancer Survivors in China 8 9 Running title: Health Literacy and QOL in Cancer 10 11 12 13 Authors full name, highest academic degrees and institutional affiliations: 14 15 Juan Xia1, PhD; Qing-Long Deng1, Rui Yan1, Peng Wu1, Ren-Ren Yang1, Bing-hui Lv1, Ji-Wei Wang1, 16 17 PhD; Jin-Ming Yu1, PhD 18 For peer review only 19 1, Institute of Clinical Epidemiology, Key Laboratory of Public Health Safety, Ministry of Education, 20 21 Key Lab of Health Technology Assessment of Ministry of Health, School of Public Health, Fudan 22 23 University 24 25 26 27 Information for Corresponding Author: 28 29 Jin-Ming Yu, School of Public Health, Fudan University, 130 Dong-An Road, Shanghai, China, 200032. 30 31 Tel: (021) 54237447; Fax: (021)54237868; E-mail: [email protected]. 32 33 34 Ji-Wei Wang, School of Public Health, Fudan University, 130 Dong-An Road, Shanghai, China, 200032. 35 36 Tel: (021)54237898; Fax: (021)54237898; E-mail: [email protected].
37 http://bmjopen.bmj.com/ 38 39 40 Manuscript word count: 2385 words 41 42 Numbers of references: 32 43 44 Numbers of tables and figures: 3 tables and 2 figure
45 on September 23, 2021 by guest. Protected copyright. 46 Abstract word count: 262 words 47 48 Key words: Health Literacy; 3 Brief Questions; Quality of Life; Cancer Survivors 49 50 51 52 Conflict of interest statement: The authors declare no potential conflicts of interest. 53 54 55 56 57 58 59 60 1
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 18
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Health Literacy and Quality of Life Among Cancer Survivors in China 5 6 7 8 9 Objective To evaluate the association between Health Literacy (HL) and Quality of Life (QOL) among Cancer Survivors 10 11 12 in China. 13 14 15 Design Cross-sectional observational study in China. 16 17 18 Setting and participants Cross-sectionalFor peer observational review study of 4713 cancer only survivors who were older than 18 years and 19 20 had cancers come from the Shanghai Cancer Rehabilitation Club. Participants were enrolled and completed questionnaires 21 22 23 between May and July 2017. 24 25 26 Main measures We assessed participants’ HL by using 3 established screening questions. Participants were excluded if 27 28 they did not complete at least 1 HL question. Item Response Theory (IRT) was used to evaluate the existing measure of 29 30 31 health literacy. Scores were summed, and participants were categorized as low HL if their total score was greater than 10 32 33 34 and adequate HL if it was 10 or lower. 35 36 Results Of the 4173 participants surveyed, 4610 responded (97.8% response rate). IRT scaling parameters of the health
37 http://bmjopen.bmj.com/ 38 39 literacy measure found items had a good range of discriminating and difficulty. Of 4589 included responders, 159 (3.5%) 40 41 42 had low HL. After adjusting for participants’ sociodemographic characteristics, social support, treatment regimen, and 43 44 years with cancers, for each 1-point decrement in HL score, the QOL score increased by 2.07 (P<.0001). Cancer survivors
45 on September 23, 2021 by guest. Protected copyright. 46 47 with low HL were less likely than survivors with adequate HL to achieve a better QOL. In logistic regression, low HL was 48 49 50 independently associated with poor QOL (adjusted odds ratio, 2.81 [95% confidence interval], 1.94-4.06 ; P<.001). 51 52 53 Conclusions Among cancer survivors in Shanghai Cancer Rehabilitation Club, low HL was independently associated with 54 55 56 poor QOL. 57 58 Key words: Health Literacy; 3 Brief Questions; Quality of Life; Cancer Survivors 59 60 2
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 18 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Article summary 5 6 7 Cancer survivors with adequate Health Literacy had nearly 3 times the odds of having a better Quality of Life than 8 9 cancer survivors with inadequate Health Literacy. 10 11 12 Improving cancer survivors' health literacy might be an important targets for improving their QOL. 13 14 15 Strengths and limitations of this study 16 17 18 To the best of our knowledge,For thispeer is the first studyreview using 3 brief screeningonly questions to evaluate Chinese cancer 19 20 survivors’ health literacy instead of the complex and long questions which are not suitable to use in clinical routine 21 22 23 practice, and to explore its relationship with quality of life. 24 25 26 Item response theory was used to evaluate whether a summed unidimensional health literacy scale could be used. 27 28 Causal inferences could not be allowed due to the cross-sectional design. 29 30 31 32 33 34 Introduction 35 36 Health literacy (HL) is an evolving concept. As defined by the National Library of Medicine(1), HL is “the degree to
37 http://bmjopen.bmj.com/ 38 39 which individuals have the capacity to obtain, process, and understand basic health information and services needed to 40 41 42 make appropriate health decisions.” It means more than simply the ability to “read pamphlets”, “make appointments”, 43 44 “understand food labels” or “comply with prescribed actions” from a doctor(2). Higher levers of HL within populations
45 on September 23, 2021 by guest. Protected copyright. 46 47 yield social benefits(3, 4), and HL is an important factor in ensuring significant health outcomes(5, 6). 48 49 50 Cancer is one of the leading causes of morbidity and mortality worldwide(7). The economic impact of cancer is 51 52 53 significant and increasing. Quality of Life (QOL) has been increasingly used as a comprehensive health indicator in clinical 54 55 56 treatment and interventions(8). Little is known about the association between HL and QOL among cancer survivors. 57 58 Isolating the independent contribution of HL toward cancer survivors’ QOL would have an important clinical and public 59 60 3
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 18
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 health implications. A growing body of research measured HL with complex and long questions which are not suitable to 5 6 7 use in clinical routine practice (9). Therefore, the purpose of this study was to evaluate the association between HL and 8 9 QOL among cancer survivors in Shanghai Cancer Rehabilitation Club using 3 brief screening questions. 10 11 12 METHODS 13 14 15 Patient and Public Involvement 16 17 18 All participants were enrolledFor in Shanghaipeer Cancer review Rehabilitation Club, only a nongovernmental self-help mutual aid 19 20 organization that contains of 20 members of the branch offices, 175 community block group, and more than 13,000 21 22 23 members. The study covered 16 districts in Shanghai (Huangpu, Pudong, Xuhui, Changning, Putuo, Hongkou, Yangpu, 24 25 26 Minhang, Baoshan, Jiading, Jinshan, Songjiang, Qingpu, Fengxian, Chongming and Jingan). 27 28 Participants were eligible if they were older than 18 years, diagnosed with cancers, cured or uncured, with or without 29 30 31 complications. Participants had to have read ability. We excluded participants with any documented billing diagnosis of 32 33 34 psychotic disorder, dementia, or blindness (conditions that may interfere with accurate health literacy measurement). 35 36 Between May and July 2017, investigators were trained and field investigation was conducted. Participants were asked
37 http://bmjopen.bmj.com/ 38 39 to participant in a survey and complete a questionnaire survey and were offered a box of eggs for their participation. 40 41 42 Informed consent was obtained from survivors before enrollment. The protocol was approved by the committee of Public 43 44 Health School of Fudan University (protocol number IRB # 2017-05-0621).
45 on September 23, 2021 by guest. Protected copyright. 46 47 Measures 48 49 50 HL was assessed using 3 established screening questions and categorized as adequate or inadequate (10-13). The 51 52 53 screening questions, respectively, were Item1: “How often do you have someone help you read hospital materials?”, Item 54 55 56 2: “How confident are you filling out forms by yourself” and Item 3: “How often do you have problems learning about 57 58 your medical condition because of difficulty reading hospital materials?”. Each question was scored by participants on a 59 60 4
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 18 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5-point scale. HL was evaluated as a continuous and dichotomous variable. Based on prior literature(9), scores were 5 6 7 summed, and participants were categorized as low HL if their total score was greater than 10 and adequate HL if it was 10 8 9 or lower. 10 11 12 QOL was assessed using the simplified Chinese version of the quality-of-life questionnaire-core 30 items (EORTC QLQ- 13 14 15 30), which incorporates nine multi-item scales(14): five functional scales (physical, role, cognitive, emotional, and social 16 17 18 functioning), three symptomFor scales (fatigue, peer pain, and review nausea/vomiting), onlyand a global health status/QOL scale; and six 19 20 single-item scales (dyspnea, insomnia, appetite loss, constipation, diarrhea, and financial difficulties). The psychometric 21 22 23 properties of the questionnaire had been previously evaluated(15). Scale scores were calculated by averaging items within 24 25 26 scales and transforming average scores linearly. All of the scales range in score from 0 to100. A high score for a functional 27 28 scale represents a high/healthy level of functioning whereas a high score for a symptom scale or item represents a high 29 30 31 level of symptomatology or problems. For more details on the scoring procedures see the EORTC QLQ-C30 Scoring 32 33 34 Manual(16). Participants were classified as having better QOL if their scores was higher than 50 and poor QOL if it was 35 36 lower than 50.
37 http://bmjopen.bmj.com/ 38 39 Other variables 40 41 42 Covariates collected were age, sex, race/ethnicity, socioeconomic status, self-reported education level, marital status 43 44 (married vs single or divorced/widowed), insurance status, years with cancer, smoking habit, alcohol use, excise, treatment
45 on September 23, 2021 by guest. Protected copyright. 46 47 regimen, district, BMI and history of coexisting illnesses. Because coexisting illnesses may affect survivors QOL, we 48 49 50 measured coexisting illnesses conditions by asking cancer survivors whether they had ever been told by a physician that 51 52 53 they had chronic diseases, including hypertension, hyperlipemia, hyperuricemia, diabetes, heart disease, stroke, respiratory 54 55 56 disease, digestive system disease and skeleton system diseases. All covariates were determined at the time of the survey. 57 58 Statistical Analysis 59 60 5
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 18
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 To determine whether a summed unidimensional health literacy scale could be used, item response theory (IRT) was 5 6 7 used to evaluate the item parameter (discrimination parameter, threshold parameter), item characteristic curve (ICC) and 8 9 item information curve (IIF). For the items were ordered responses, the Graded Response Model (GRM) was used. For the 10 11 12 participants answering only 1 or 2 of the 3 questions, the score for those questions was multiplied by 3 and 1.5, respectively 13 14 15 (9). 16
17 2 18 Baseline characteristics wereFor compared peer across levels review of HL using 휒 testonly for categorical variables and Wilcoxon test 19 20 for the non-normal continuous variables. Linear regression models were used to estimate the association between HL score 21 22 23 and QOL after controlling for differences in participants’ characteristics, including age, sex, race/ethnicity, marriage, 24 25 26 education, income, insurance status, years with cancer, number of chronic disease, smoking habit, alcohol use, excise, 27 28 treatment regimen, district and BMI. Logistic regression models were used to measure the independent relationship 29 30 31 between HL and each of the QOL scales, while adjusting for other potentially confounding survivor characteristics. 32 33 34 The IRT analyses were conducted using the MULTILOG version 7.03. Other statistical analyses were performed with 35 36 SAS version 9.4 for each analysis, the null hypothesis was evaluated at a 2-sided significance levels of 0.05.
37 http://bmjopen.bmj.com/ 38 39 40 41 42 Results 43 44 Among 4713 cancer survivors surveyed, 4610 responded, for a 97.81% response rate. For 4589 of the 4610 participants,
45 on September 23, 2021 by guest. Protected copyright. 46 47 at least 1 HL question was available in the survey; these participants composed our study sample. 48 49 50 The summed HL scale was measured by using all 3 HL questions. The correlations of single items to the total were 0.73, 51 52 53 0.70, and 0.75 for questions 1 to 3, respectively. Figure 1 shows the IRT models and test information for the total sample 54 55 56 and illustrates the implications of the item parameters for the response probabilities for different levels of health literacy. 57 58 All items had high discrimination parameters, except for Item 3, ‘How often do you have problems learning about your 59 60 6
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 18 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 medical condition because of difficulty reading hospital materials’, which had a moderate discrimination (a=0.88) but still 5 6 7 satisfy the condition of a>0.75. The difficulty parameters (Thresholdi) increased monotonically across rating scale 8 9 categories for all items, ranging from -1.81 to 3.94, indicating that the 3 HL questions discriminate well. Among the 4589 10 11 12 participants in the study, the mean HL score was 6.3 (range, 3-15). As table 1 shows, one hundred fifty-nine participants 13 14 15 (3.5%) had low HL (HL score, 11-15). Participants with low HL were more likely than participants with adequate HL to 16 17 18 be older, to have received onlyFor some middlepeer school education review or less, and toonly have had little excise. 19 20 Table 2 presents the bivariate relationships between predictors of QOL and cancer survivors’ global health status. The 21 22 23 HL score, age, sex, marriage, years with cancer, number of the chronic disease, smoking habit, alcohol use, excise, district 24 25 26 and BMI were all associated with QL. After adjustment for other potentially confounding factors, marriage, years with 27 28 cancer, and smoking habit were no significant difference. For each 1-point decrement in HL score, the QL value increased 29 30 31 by 2.07 (P<.001). 32 33 34 Cancer survivors with adequate HL were more likely than cancer survivors with low HL to have a high score for a 35 36 functional scale whereas a low level of symptomatology. According to the recommendation(17), a difference on the 0–100
37 http://bmjopen.bmj.com/ 38 39 scale of more than 5 points were considered as a clinically important difference. As can be seen from the data in Figure 2 40 41 42 that there had a clinically significant difference between survivors with adequate HL and survivors with low HL, in terms 43 44 of global health status (difference of adequate HL vs low HL [diff], 9.05), role function (diff: 5.2), emotional function (diff:
45 on September 23, 2021 by guest. Protected copyright. 46 47 5.17), cognitive function (diff: 7.13), social function (diff: 7.49), insomnia (diff: -8.8), and financial difficulties (diff: - 48 49 50 8.71). 51 52 53 Eighty-three percent of cancer survivors with adequate HL and 65% of cancer survivors with low HL reported that they 54 55 56 had a good global health status (unadjusted odds ratio [OR], 2.75; 95% confidence interval [CI], 1.97-3.84; P<.001). After 57 58 confounders were adjusted, cancer survivors with adequate HL were more likely to report a better global health status 59 60 7
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 18
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 (adjusted OR, 2.81; 95% CI, 1.94-4.06; P<.001) (Table 3). The extent of the associations between HL and other scales of 5 6 7 QOL, including four functional scales (physical, emotional, cognitive and social functioning), three symptom scales 8 9 (fatigue, pain, and nausea/vomiting), and three single-item (insomnia, appetite loss, and financial difficulties), was similar 10 11 12 to that of global health status and all had a statistical significance (Table 3). 13 14 15 Discussion 16 17 18 Our study demonstrates thatFor inadequate peer HL as assessed review by 3 brief screening only questions was present in less than 1 in 25 19 20 cancer survivors in Shanghai Cancer Rehabilitation Club. And inadequate HL was an independent predictor of poor QOL 21 22 23 and was associated with a lower level of functioning and higher level of symptomatology or problems. The association 24 25 26 between HL and QOL that we observed is significant from a clinical and public health perspective. These findings highlight 27 28 a potential target for interventions to improve the overall quality of life for cancer survivors. 29 30 31 To our knowledge, no prior study had demonstrated the association between HL and QOL among cancer survivors using 32 33 34 the 3 brief screening questions. Prior research studies have used more complex and extensive questionnaires to measure 35 36 HL, such as the Rapid Estimate of Adult Literacy in Medicine(18, 19), Cancer Health Literacy Test-30 (CHLT-30)(20),
37 http://bmjopen.bmj.com/ 38 39 the Short-Form Test of Functional Health Literacy in Adults (S-TOFHLA)(5, 21, 22), et al. which are also impractical for 40 41 42 use in busy clinical settings. Despite differences in the nature of HL assessment techniques, our results are consistent with 43 44 those of previous studies that identified inadequate HL as a risk factor for QOL among cancer survivors.(19, 23, 24)
45 on September 23, 2021 by guest. Protected copyright. 46 47 Our results showing a higher adequate HL level are in contrast to those from other studies(25, 26) which found a lower 48 49 50 adequate HL in Chinese population. This might because the participants we surveyed were all come from the Cancer 51 52 53 Rehabilitation Club in Shanghai. The club(27) firmly “anti-cancer groups, beyond life”, and organizing a variety of 54 55 56 rehabilitation activities to make patients feel full confidence of their rehabilitation. They spread new concept of treatment 57 58 and health education mode their own health, such as “group psychotherapy”, “create a new life”, “qigong physical exercise”, 59 60 8
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 18 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 “five-full philosophy”, “participation in management”, et al. Through this process, participants in the club will have their 5 6 7 HL level gradually improved. This might be the reason why our results were higher than other studies. 8 9 We chose QOL as the primary outcome of interest were because it can fully reflect the feeling and physical recovery 10 11 12 condition of the patients and has been increasingly used as a comprehensive health indicator in clinical treatment and 13 14 15 interventions(8, 28). QOL can be used as one of the indicators of efficacy of different therapeutic measures. The 16 17 18 combination of QOL and clinicalFor curative peer effect observation review index can help only to choose the treatment that is more suitable 19 20 for patients. To evaluate the difference of QOL before and after treatment can help doctors have a whole understanding 21 22 23 about the patients’ true feelings and health, so that they can in different stages of different measures. Studies have 24 25 26 demonstrated that there is a negative relationship between QOL and HL(19, 24). Inadequate health literacy is independently 27 28 associated with poor QOL and higher rates of (29). Consistent with these researches, our study showed that cancer survivors 29 30 31 with adequate HL had nearly 3 times the odds of having a better QOL than cancer survivors with inadequate HL. 32 33 34 Form the public health perspective, HL is an important factor in ensuring significant health outcomes.(30-32) Inadequate 35 36 HL may contribute to the disproportionate burden of cancer-related problems among disadvantaged populations. The
37 http://bmjopen.bmj.com/ 38 39 United Nations ECOSOC Ministerial Declaration of 2009 provided a clear mandate for action “call for the development 40 41 42 of appropriate action plans to promote health literacy.”(2) The 9th Global Conference was held in Shanghai, China (2016) 43 44 and addressed the determinants of health through good governance, health cities, health literacy and social mobilization.(2)
45 on September 23, 2021 by guest. Protected copyright. 46 47 Therefore, improving cancer survivors' health literacy is an important targets for improving their QOL. 48 49 50 Several limitations should be considered in the interpretation of our findings. First, it was a cross-sectional study and 51 52 53 didn’t allow us to make a causal inference whether low HL was causally associated with poor QOL. Second, we didn’t use 54 55 56 the questionnaire of Chinese residents health literacy monitoring and other complex tool to evaluate the HL, and the 3 brief 57 58 questions may reflect other constructs, but the 3 brief questions can easily be incorporated into the clinical routine work 59 60 9
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 18
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 and are useful to identify high-risk patients. Third, this study might only have a good representativeness of the cancer 5 6 7 survivors in Shanghai Cancer Rehabilitation Club and may not be generalizable to non-members of Cancer Rehabilitation 8 9 Club. However, as discussed above, if the effect of inadequate HL would be weaken in this participants, it may be greater 10 11 12 in different populations. 13 14 15 Conclusions 16 17 18 In summary, inadequate healthFor literacy peer as evaluated review by 3 brief screening questionsonly is independently associated with poor 19 20 QOL in cancer survivors. Efforts should focus on developing and evaluating interventions to improve QOL among cancer 21 22 23 survivors with inadequate HL. 24 25 26 DECLARATIONS 27 28 Ethics approval and consent to participate: Informed consent was obtained from all participants before enrollment. The 29 30 31 protocol was approved by the committee of Public Health School of Fudan University (protocol number IRB # 2017-05- 32 33 34 0621). 35 36 Consent for publication: Not applicable
37 http://bmjopen.bmj.com/ 38 39 Availability of data and material: Please contact author for data requests. 40 41 42 Competing interests: The authors declare that they have no competing interests. 43 44 Funding: Not applicable
45 on September 23, 2021 by guest. Protected copyright. 46 47 Authors' contributions: Juan Xia carried out the field investigation, participated in the design of the study, performed the 48 49 50 statistical analysis and drafted the manuscript. Qing-Long Deng, Rui Yan, Peng Wu, Ren-Ren Yang and Bing-hui Lv 51 52 53 participated in the field investigation and data collection. Ji-Wei Wang and Jin-Ming Yu designed and coordinated the 54 55 56 study and revised the manuscript. All authors read and approved the final manuscript. 57 58 Acknowledgements: We wish to thank Furong Tang, Donghui Yu, Mengxue Yin and some other volunteers who 59 60 10
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 18 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 contributed to this survey. We also wish to thank the thousands of cancer survivors who were willing to complete the 5 6 7 survey. 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 11
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 18
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Reference 5 1. Kindig, D. A., Panzer, A. M., & Nielsen-Bohlman, L. editors. Health Literacy: A Prescription to End Confusion. 6 7 National Academies Press 15, 2004:389-395. 8 2. World Health Organization. http://www.who.int/healthpromotion/conferences/9gchp/health-literacy/en/ 9 Accessed February 1, 2018. 10 11 3. Eichler K, Wieser S, Brügger U. The costs of limited health literacy: a systematic review. International Journal 12 of Public Health. 2009; 54, 313. 13 4. Baker DW, Parker RM, Williams MV, et al. The relationship of patient reading ability to self-reported health and 14 15 use of health services. American Journal of Public Health..1997; 87, 1027. 16 5. Schillinger D, Grumbach K, Piette J, et al. Association of Health Literacy With Diabetes Outcomes. JAMA. 17 18 2002;288(4):475–482.For peer review only 19 6. Weiss BD, Hart G, Mcgee DL, et al. Health status of illiterate adults: relation between literacy and health status 20 among persons with low literacy skills. Journal of the American Board of Family Practice.1992; 5, 257. 21 22 7. International Agency for Research on Cancer. GLOBOCAN 2012: Estimated Cancer Incidence, Mortality and 23 Prevalence Worldwide in 2012. http://globocan.iarc.fr/ Accessed February 1, 2018. 24 8. Liu Z, Yang J. Health related quality of life: is it another comprehensive evaluation indicator of Chinese medicine 25 26 on acquired immune deficiency syndrome treatment? Journal of Traditional Chinese Medicine. 2015; 35(5):600-5. 27 9. Peterson PN, Shetterly SM, Clarke CL,et al. Health Literacy and Outcomes Among Patients With Heart Failure. 28 JAMA. 2011;305(16):1695–1701. 29 30 10. Chew LD, Griffin JM, Partin MR, et al. Validation of screening questions for limited health literacy in a large VA 31 outpatient population. J Gen Intern Med. 2008; 23(5), 561-566. 32 33 11. Wallace LS, Rogers ES, Roskos SE, et al. Brief report: screening items to identify patients with limited health 34 literacy skills. J Gen Intern Med. 2006;21(8):874-7. 35 12. Wallace LS, Cassada DC, Rogers ES, et al. Can screening items identify surgery patients at risk of limited health 36 literacy? J Surg Res. 2007;140(2):208-13.
37 http://bmjopen.bmj.com/ 38 13. Chew LD, Bradley KA, Boyko EJ. Brief questions to identify patients with inadequate health literacy. Fam Med. 39 2004;36(8):588-94. 40 41 14. Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization for Research and Treatment of Cancer 42 QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst. 43 1993;85(5):365-76. 44 15. Wang JW. Tang Z, Yu JM, et al. QLQ-BR23 and EORTC+QLQ-C30 for measurement of the impact of
45 on September 23, 2021 by guest. Protected copyright. 46 rehabilitation exercise on quality of life in breast cancer patients. Fudan Univ J Med Sci, 2015. 47 16. Fayers PM, Aaronson NK, Bjordal K. The EORTC QLQ-C30 scoring manual. 2001. 48 49 17. Osoba D, Rodrigues G, Myles J. Interpreting the significance of changes in health-related quality-of-life scores. 50 J Clin Oncol. 1998;16(1):139-44. 51 52 18. Davis TC, Long SW, Jackson RH, et al. Rapid estimate of adult literacy in medicine: a shortened screening 53 instrument. Fam Med. 1993;25(6):391-5. 54 19. Song L, Mishel M, Bensen JT, et al. How does health literacy affect quality of life among men with newly 55 56 diagnosed clinically localized prostate cancer? Findings from the North Carolina-Louisiana Prostate Cancer 57 Project (PCaP). Cancer. 2012;118(15):3842-51. 58 20. Echeverri M, Anderson D, Nápoles AM. Cancer Health Literacy Test-30-Spanish (CHLT-30-DKspa), a New 59 60 Spanish-Language Version of the Cancer Health Literacy Test (CHLT-30) for Spanish-Speaking Latinos. J Health 12
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 18 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from Commun. 2016;21 Suppl 1:69-78. 4 5 21. Yehle KS, Plake KS, Nguyen P, et al. Health-Related Quality of Life in Heart Failure Patients With Varying 6 Levels of Health Literacy Receiving Telemedicine and Standardized Education. Home Healthc Now. 7 2016;34(5):267-72. 8 9 22. Baker DW, Williams MV, Parker RM, et al. Development of a brief test to measure functional health literacy. 10 Patient Educ Couns. 1999;38(1):33-42. 11 23. Husson O, Mols F, Fransen MP, et al. Low subjective health literacy is associated with adverse health behaviors 12 13 and worse health-related quality of life among colorectal cancer survivors: results from the profiles registry. 14 Psychooncology. 2015;24(4):478-86. 15 16 24. Halverson JL, Martinez-Donate AP, Palta M, et al. Health Literacy and Health-Related Quality of Life Among a 17 Population-Based Sample of Cancer Patients. J Health Commun. 2015;20(11):1320-9. 18 25. Yao H, Shi Q, Li Y.For The Current peer Status of Health review Literacy in China. only China Population Today. 2016; 41-41. 19 20 26. Huang L, Peng S, University FM, et al. The Status and Influencing Factors of Health Literacy among Cancer 21 Patients. Nursing Journal of Chinese Peoples Liberation Army. 2016. 22 27. http://www.shcrc.cn/webs/intro.aspx. Published 2012. Accessed February 1, 2018. 23 24 28. Guyatt GH, Feeny DH, Patrick DL. Measuring Health-Related Quality-of-Life. Ann Intern Med. 1993; 25 118(8):622-9. 26 29. Wang CL, Kane RL, Xu DJ, et al. Health literacy as a moderator of health-related quality of life responses to 27 28 chronic disease among Chinese rural women. BMC Womens Health. 2015;15:34. 29 30. Protheroe J, Wolf MS, Lee A. Health literacy and health outcomes. Health Literacy in Context: International 30 Perspectives. 2010;57-74. 31 32 31. Halverson JL, Martinez-Donate AP, Palta M, et al. Health Literacy and Health-Related Quality of Life Among a 33 Population-Based Sample of Cancer Patients. J Health Commun. 2015;20(11):1320-9. 34 35 32. Aaby A, Friis K, Christensen B, et al. Health literacy is associated with health behaviour and self-reported health: 36 A large population-based study in individuals with cardiovascular disease. Eur J Prev Cardiol. 2017;24(17):1880-
37 1888. http://bmjopen.bmj.com/ 38 39 40 41 Figure Title: 42 43 44 Figure 1. Item Characteristic Curve and Test Information Curve for the 3 established screening questions
45 on September 23, 2021 by guest. Protected copyright. 46 47 Figure 2. Unadjusted QOL scores for cancer survivors with inadequate and adequate health literacy 48 49 50 51 52 53 54 55 56 57 58 59 60 13
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 18
1 2 3 Table 1. Characteristics of Respondent stratified by Health Literacy Level* BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 Level of Health Literacy 6 Characteristics Total (N=4589) Adequate (n=4430) Inadequate (n=159) P value 7 8 Age, M (Q1,Q3), y 62.0 (57.04,66.53) 62.0 (57.00,66.53) 62.52 (58.37,68.39) 0.0485 9 Years with cancers, M (Q1,Q3) 6.09 (3.50,10.50) 6.09 (3.50,10.50) 5.92 (3.50,10.58) 0.6258 10 Sex, No.(%) 0.9424 11 12 Male 1057 (23.03) 1020 (96.50) 37 (3.50) 13 Female 3532 (76.97) 3410 (96.55) 122 (3.45) 14 Race 0.8644 15 16 Hanzu 4544 (99.28) 4386 (96.52) 158 (3.48) 17 Marriage 0.2192 18 single For peer129 (2.81) review121 (93.80) only8 (6.20) 19 20 married/ cohabitation 3989 (86.93) 3853 (96.61) 136 (3.41) 21 divorced/widowed/ separated 471 (10.226) 456 (96.82) 15 (3.18) 22 Annual income, ¥ 0.0534 23 24 <5000 4134 (91.89) 3987 (96.44) 147 (3.56) 25 ≥5000 365 (8.11) 359 (98.36) 6 (1.64) 26 27 Family income 0.8296 28 <5000 3086 (70.07) 2979 (96.53) 107 (3.47) 29 ≥5000 1318 (29.93) 1274 (96.66) 44 (3.34) 30 31 Education 0.0003 32 Middle school or less 4021 (87.62) 3867 (96.17) 154 (3.83) 33 High school or college 568 (12.38) 563 (99.12) 5 (0.88) 34 35 Insurance status 0.8159 36 Uninsured 645 (15.55) 623 (96.59) 22 (3.41)
37 http://bmjopen.bmj.com/ Medicare 3504 (84.45) 3378 (96.40) 126 (3.60) 38 39 Treatment regimen 0.2753 40 Chemotherapy 661 (14.44) 641 (96.97) 20 (3.03) 41 Radiotherapy 74 (1.62) 68 (91.89) 6 (8.11) 42 43 Operation 931 (20.34) 902 (96.89) 29 (3.11) 44 Chemotherapy & Radiotherapy 162 (3.54) 155 (95.68) 7 (4.32)
45 on September 23, 2021 by guest. Protected copyright. 46 Chemotherapy & Radiotherapy 897 (19.59) 870 (96.99) 27 (3.01) 47 & Operation 48 Chemotherapy & Operation 1658 (36.22) 1594 (96.14) 64 (3.86) 49 50 Biotherapy 69 (1.51) 66 (95.65) 3 (4.35) 51 Number of Chronic disease 0.3564 52 0 1238 (26.98) 1203 (97.17) 35 (2.83) 53 54 1-3 2735 (59.60) 2634 (96.31) 101 (3.69) 55 >3 616 (13.42) 593 (96.27) 23 (3.73) 56 Smoking habit 0.5284 57 58 Never 3841 (83.70) 3711 (96.62) 130 (3.38) 59 Former 598 (13.03) 573 (95.82) 25 (4.18) 60 14
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 18 BMJ Open
1 2 3 Current 150 (3.27) 146 (97.33) 4 (2.67) BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 Current alcohol use 0.0637 6 None 4264 (92.92) 4113 (96.46) 151 (3.54) 7 Light to moderate 215 (4.69) 213 (99.07) 2 (0.93) 8 9 Heavy 110 (2.40) 104 (94.55) 6 (5.45) 10 Body mass index# 0.4593 11 <18.5 233 (5.16) 229 (98.28) 4 (1.72) 12 13 18.5-23.9 2342 (51.85) 2258 (96.41) 84 (3.59) 14 24.0-27.9 1546 (34.23) 1492 (96.51) 54 (3.49) 15 16 ≥28.0 396 (8.77) 380 (95.96) 16 (4.04) 17 Excise 0.0264 18 No For peer1038 (22.64) review991 (95.47) only47 (4.53) 19 20 1-4 times per week 2493 (54.37) 2406 (96.51) 87 (3.49) 21 ≥5 times per week 1054 (22.99) 1029 (97.63) 25 (2.37) 22 * Data are given as percentages unless otherwise indicated. 23 24 # Calculated as weight in kilograms divided by the square of height in meters. 25 26 Table 2. Relationship Between Survivors Characteristics and QOL 27 28 Unadjusted Adjusted* Predictor 29 Coefficientτ P Value Coefficientτ P Value 30 31 HL score -2.11 <0.0001 -2.07 <0.0001 32 Age -0.25 <0.0001 -0.16 0.0053 33 Sex, female -3.92 <0.0001 -5.02 0.0002 34 35 Race/ethnicity 2.10 0.5854 6.41 0.1819 36 Marriage -3.01 0.0051 -2.14 0.0710
37 http://bmjopen.bmj.com/ Education -0.28 0.6316 -1.38 0.0319 38 39 Annual income 0.42 0.7664 0.81 0.6044 40 Family income -0.03 0.9694 0.53 0.5521 41 42 Insurance status -1.67 0.1263 -0.09 0.9346 43 Years with cancer (per year) -0.23 0.0002 -0.04 0.5401 44 Number of Chronic disease -5.79 <0.0001 -4.87 <0.0001
45 on September 23, 2021 by guest. Protected copyright. 46 Smoking habit 1.35 0.015 -1.23 0.1210 47 Alcohol intake 3.38 <0.0001 2.07 0.0240 48 Excise 4.07 <0.0001 3.15 <0.0001 49 50 Treatment regimen -0.004 0.985 0.16 0.4474 51 District 0.63 <0.0001 0.48 <0.0001 52 BMI 1.14 0.0305 2.18 <0.0001 53 54 * Adjusted for age, sex, race, marriage, income, education, insurance, years with cancer, treatment regimen, district, 55 number of chronic disease, smoking habit, alcohol use, excise and BMI. 56 τAll coefficients correspond to a change in QOL score for unit change of each covariate. 57 58 59 60 15
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 18
1 2 3 Table 3. Adjusted Odds Ratio of QOL for cancer survivors with inadequate vs adequate health literacy* BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 QOL Survivors with poor QOL Odds Ratio (95%CI) P value 6 Global health status (QL) 788 2.81 (1.94-4.06) <0.0001 7 Physical function (PF) 83 2.84 (1.18-6.87) 0.0205 8 9 Role function (RF) 53 2.70 (0.94-7.80) 0.0664 10 Emotional function (EF) 77 5.01 (2.37-10.58) <0.0001 11 12 Cognitive function (CF) 128 4.07 (2.18-7.61) <0.0001 13 Social function (SF) 250 3.59 (2.19-5.88) <0.0001 14 Fatigue (FA) 425 2.63 (1.68-4.12) <0.0001 15 16 Nausea/Vomiting (NV) 52 5.39 (2.17-13.38) 0.0003 17 Pain (PA) 296 2.37 (1.41-3.97) 0.0011 18 Dyspnea (DY) For 152peer review1.86 only (0.88-3.94) 0.106 19 20 Insomnia (SL) 475 2.41 (1.56-3.74) <0.0001 21 Appetite loss (AP) 91 3.84(1.84-8.03) 0.0003 22 Constipation (CO) 188 1.06 (0.45-2.46) 0.8995 23 24 Diarrhea (DI) 114 2.11 (0.95-4.71) 0.0679 25 Financial difficulties (FI) 747 2.73 (1.88-3.97) <0.0001 26 27 *Adjusted for age, sex, race, marriage, income, education, insurance, years with cancer, treatment regimen, district, number 28 of chronic disease, smoking habit, alcohol use, excise and BMI 29 30 31 32 33 34 35 36
37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 16
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 18 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 Item 1 Item 2 6 Slope=1.49 Slope= 2.21 7 Threshold1= -1.03 Threshold1= -0.15 Threshold2= 0.81 Threshold2= 1.21 8 Threshold3= 2.48 Threshold3= 2.02 9 Threshold4= 3.80 Threshold4= 2.48 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 Item 3 24 Slope= 0.88 25 Threshold1= -1.81 Threshold2= -0.05 26 Threshold3= 2.47 27 Threshold4= 3.94 28 29 30 31 32 33 34 35 36
37 http://bmjopen.bmj.com/ 38 39 Figure 1. Item Characteristic Curve and Test Information Curve for the 3 established screening questions 40 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 18
1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 2. Unadjusted QOL scores for cancer survivors with inadequate and adequate health literacy 33 34 35 36
37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from
Health Literacy and Quality of Life among Cancer Survivors in China
Journal: BMJ Open ManuscriptFor ID peerbmjopen-2018-028458.R1 review only Article Type: Original research
Date Submitted by the 19-Jul-2019 Author:
Complete List of Authors: Xia, Juan; Fudan University, school of public health Wu, Peng; Fudan University, school of public health Deng, Qinglong; Fudan University, school of public health Yan, Rui; Fudan University, school of public health Yang, Renren; Fudan University, school of public health Lv, Binghui; Fudan University, school of public health Wang, Jiwei; Fudan University, Fudan university school of public health Yu, Jinming; Fudan University, school of public health
Primary Subject Public health Heading:
Secondary Subject Heading: Oncology
Keywords: Health Literacy, 3 Brief Questions, Quality of Life, Cancer Survivors http://bmjopen.bmj.com/
on September 23, 2021 by guest. Protected copyright.
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 21 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Title Page 5 6 Title: Health Literacy and Quality of Life among Cancer Survivors in China 7 8 Running title: Health Literacy and QOL in Cancer 9 10 11 12 Authors full name, highest academic degrees and institutional affiliations: 13 14 Juan Xia1, PhD, Peng Wu1, Qing-Long Deng1, Rui Yan1, Ren-Ren Yang1, Bing-hui Lv1, Ji-Wei Wang1, 15 16 PhD, Jin-Ming Yu1, PhD 17 18 1, Institute of ClinicalFor Epidemiology, peer Key Laboratoryreview of Public Healthonly Safety, Ministry of Education, 19 20 Key Lab of Health Technology Assessment of Ministry of Health, School of Public Health, Fudan 21 22 University 23 24 Juan Xia and Peng Wu have contributed equally to this work. 25 26 27 28 Information for Corresponding Author: 29 30 Jin-Ming Yu, School of Public Health, Fudan University, 130 Dong-An Road, Shanghai, China, 200032. 31 32 Tel: (021) 54237447; Fax: (021)54237868; E-mail: [email protected]. 33 34 Ji-Wei Wang, School of Public Health, Fudan University, 130 Dong-An Road, Shanghai, China, 200032. 35 36 Tel: (021)54237898; Fax: (021)54237898; E-mail: [email protected].
37 http://bmjopen.bmj.com/ 38 39 40 Manuscript word count: 2858 words 41 42 Numbers of references: 39 43 44 Numbers of tables and figures: 3 tables and 2 figure
45 on September 23, 2021 by guest. Protected copyright. 46 Abstract word count: 290 words 47 48 Key words: Health Literacy; 3 Brief Questions; Quality of Life; Cancer Survivors 49 50 51 52 53 Conflict of interest statement: The authors declare no potential conflicts of interest. 54 55 56 57 58 59 60 1
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 21
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Health Literacy and Quality of Life among Cancer Survivors in China 5 6 7 8 Objective To evaluate the association between Health Literacy (HL) and Quality of Life (QOL) among Cancer Survivors 9 10 in China. 11 12 Design Cross-sectional observational study in China. 13 14 Setting and participants Cross-sectional observational study of 4713 cancer survivors who were older than 18 years came 15 16 from the Shanghai Cancer Rehabilitation Club. Participants were enrolled and completed questionnaires between May and 17 18 July 2017. For peer review only 19 20 Measurement HL were accessed by 3 established screening questions and QOL was evaluated using the simplified 21 22 Chinese version of the quality-of-life questionnaire-core 30 items (EORTC QLQ-30). All questionnaires were collected 23 24 through face-to-face interviews or self-administered by literate participants. Participants were excluded if they did not 25 26 complete at least 1 HL question. Baseline characteristics were compared by levels of HL using 휒2 test for categorical 27 28 variables and Wilcoxon test for the non-normal continuous variables. Item Response Theory (IRT) was used to evaluate 29 30 the existing measure of health literacy. Linear regression models and Logistic regression models were used to investigate 31 32 the association between HL and QOL. SAS 9.4 and MULTILOG 7.03 were applied for analysis. 33 34 Results Valid selection of subjects in this study were 4589. IRT scaling parameters of the health literacy measure found 35 36 items had a good range of discriminating and difficulty. Of 4589 included responders, 159 (3.5%) had low HL. After
37 http://bmjopen.bmj.com/ 38 adjusting for participants’ sociodemographic characteristics, treatment regimen, and years with cancers, for each 1-point 39 40 decrement in HL score, the QOL score increased by 2.07 (P<0.001). Cancer survivors with low HL were less likely than 41 42 those with adequate HL to achieve a better QOL. In logistic regression, low HL was independently associated with poor 43 44 QOL (adjusted odds ratio, 2.81 [95% confidence interval], 1.94-4.06; P<0.001).
45 on September 23, 2021 by guest. Protected copyright. 46 Conclusions Among cancer survivors in Shanghai Cancer Rehabilitation Club, low HL was independently associated with 47 48 poor QOL. 49 50 51 Key words: Health Literacy; 3 Brief Questions; Quality of Life; Cancer Survivors 52 53 Article summary 54 55 Cancer survivors with adequate Health Literacy had nearly 3 times the odds of having a better Quality of Life than 56 57 cancer survivors with inadequate Health Literacy. 58 59 Improving cancer survivors' health literacy might be an important targets for improving their QOL. 60 2
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 21 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Strengths and limitations of this study 5 6 To the best of our knowledge, few studies using 3 brief screening questions to evaluate Chinese cancer survivors’ 7 8 health literacy instead of the complex and long questions which are not suitable to use in clinical routine practice, and 9 10 to explore its relationship with quality of life. 11 12 Item response theory was used to evaluate whether a summed unidimensional health literacy scale could be used. 13 14 Causal inferences could not be allowed due to the cross-sectional design. 15 16 The proportion of high HL might be overestimated, and low literacy patients are often excluded from the study 17 because of illiteracy. 18 For peer review only 19 20 Introduction 21 22 Health literacy (HL) is an evolving concept. As defined by the National Library of Medicine (1), HL is “the degree to 23 24 which individuals have the capacity to obtain, process, and understand basic health information and services needed to 25 26 make appropriate health decisions.” It means more than simply the ability to “read pamphlets”, “make appointments”, 27 28 “understand food labels” or “comply with prescribed actions” from a doctor (2). Higher levers of HL within populations 29 30 yield social benefits (3, 4), and HL is an important factor in ensuring significant health outcomes (5, 6). 31 32 Cancer is one of the leading causes of morbidity and mortality worldwide (7) and the economic impact of cancer is 33 34 increasing. Despite the fact that China had a lower incidence rate of cancer compared with western counties, it increased 35 36 with a sharp slope in decades (8, 9). Meanwhile, patients nowadays are detected with advanced diagnostic techniques and
37 http://bmjopen.bmj.com/ 38 cured by modern therapeutic methods, which resulted in the extension of lifespan of cancer patients and more attention 39 40 paid to the quality of life in their survival years (10). 41 42 Quality of Life (QOL) has been increasingly used as a comprehensive health indicator in clinical treatment and 43 44 interventions (11). Studies evaluating the relationship between HL and HRQOL produced mixed results among cancer
45 on September 23, 2021 by guest. Protected copyright. 46 survivors. Researches (12-14) conducted among cancer patients were reported that HL was positively related to the QOL. 47 48 While Elizabeth A . Hahn et al. indicated that low literacy is not an independent risk factor for poorer HRQL among breast 49 50 51 cancer (15). Isolating the independent contribution of HL toward cancer survivors’ QOL would have important clinical 52 53 and public health implications. A growing body of research measured HL with complex and long questions that are not 54 55 suitable to use in clinical routine practice (16). Therefore, the purpose of this study was to evaluate the association between 56 57 HL and QOL among cancer survivors in Shanghai Cancer Rehabilitation Club using 3 brief screening questions. 58 59 60 3
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 21
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 METHODS 5 6 Design and setting 7 8 A cross-sectional study was conducted from May to July 2017. All participants were recruited from Shanghai Cancer 9 10 Rehabilitation Club, a nongovernmental self-help mutual aid organization that contains 20 members of the branch offices, 11 12 175 community block group, and more than 13,000 members. The present study covered 16 districts in Shanghai (Huangpu, 13 14 Pudong, Xuhui, Changning, Putuo, Hongkou, Yangpu, Minhang, Baoshan, Jiading, Jinshan, Songjiang, Qingpu, Fengxian, 15 16 Chongming, and Jingan). 17 18 Participants For peer review only 19 20 The subjects of our study were cancer survivors who took part in SCRC. The inclusion criteria for study enrollment were 21 22 as shown below: 1) at least 18-year-old; 2) have a pathological diagnosis of cancer; 3) able to independently participate in 23 24 the cancer rehabilitation club; 4) willingness to provide written informed consent; 5) no cognitive impairment or psychotic 25 26 disorder. The information leaflets about the content and purpose of this study and written informed consent forms were 27 28 obtained from patients who met the inclusion criteria ahead of the investigation, and a box of eggs was offered for their 29 30 participation as gifts. Investigators who were all students of Fudan University were trained and field investigation was 31 32 conducted. Questionnaires were collected through face-to-face interviews with the help of well-trained field workers or 33 34 self-administered by literate participants. 4713 cancer survivors were surveyed, 4610 got responses. Except for the 35 36 incomplete, we totally acquired 4589 valid questionnaires. The protocol was approved by the committee of the Public
37 http://bmjopen.bmj.com/ 38 Health School of Fudan University (protocol number IRB # 2017-05-0621). 39 40 Measures 41 42 HL was assessed using 3 established screening questions and categorized as adequate or inadequate (17-20). The items 43 44 included: “How often do you have someone help you read hospital materials?”, “How confident are you filling out forms
45 on September 23, 2021 by guest. Protected copyright. 46 by yourself”, and “How often do you have problems learning about your medical condition because of difficulty reading 47 48 hospital materials?”. Each question was scored by participants on a 5-point scale. HL was evaluated as a continuous and 49 50 51 dichotomous variable. Based on prior literature (16), scores were summed, and participants were categorized as low HL if 52 53 their total score was greater than 10 and adequate HL if it was 10 or lower. 54 55 QOL was assessed using the simplified Chinese version of the quality-of-life questionnaire-core 30 items (EORTC QLQ- 56 57 30), which incorporates nine multi-item scales (21): five functional scales (physical, role, cognitive, emotional, and social 58 59 functioning), three symptom scales (fatigue, pain, and nausea/vomiting), and a global health status/QOL scale; and six 60 4
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 21 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 single-item scales (dyspnea, insomnia, appetite loss, constipation, diarrhea, and financial difficulties). The psychometric 5 6 properties of the questionnaire had been previously evaluated (22). Scale scores were calculated by averaging items within 7 8 scales and transforming average scores linearly. All of the scales range in scores from 0 to100. A high score for a functional 9 10 scale represents a high/healthy level of functioning whereas a high score for a symptom scale or item represents a high 11 12 level of symptomatology or problems. For more details on the scoring procedures see the EORTC QLQ-C30 Scoring 13 14 Manual (23). Participants were classified as having better QOL if their scores were higher than 50 and poor QOL if it was 15 16 lower than 50. 17 18 Other variables For peer review only 19 20 Covariates were age, sex, race/ethnicity, socioeconomic status, education level, marital status (married vs single or 21 22 divorced/widowed), insurance status, years with cancer, smoking habit, alcohol use, physical activity, treatment regimen, 23 24 district, BMI and history of coexisting illnesses. Because coexisting illnesses may affect survivors QOL, we measured 25 26 coexisting illnesses conditions by asking cancer survivors whether they had ever been told by a physician that they had 27 28 chronic diseases, including hypertension, hyperlipemia, hyperuricemia, diabetes, heart disease, stroke, respiratory disease, 29 30 digestive system disease, and skeleton system diseases. All covariates were determined at the time of the survey. 31 32 Statistical Analysis 33 34 To determine whether a summed unidimensional health literacy scale could be used, item response theory (IRT) was 35 36 used to evaluate the item parameter (discrimination parameter, threshold parameter), item characteristic curve (ICC) and
37 http://bmjopen.bmj.com/ 38 item information curve (IIF). For the items were ordered responses, the Graded Response Model (GRM) was used. For the 39 40 participants answering only 1 or 2 of the 3 questions, the score for those questions was multiplied by 3 and 1.5, respectively 41 42 (16). 43 44 Baseline characteristics were compared across levels of HL using 휒2 test for categorical variables and the Wilcoxon
45 on September 23, 2021 by guest. Protected copyright. 46 test for the non-normal continuous variables. Linear regression models were used to estimate the association between HL 47 48 score and QOL after controlling for differences in participants’ characteristics, including age, sex, race/ethnicity, marriage, 49 50 51 education, income, insurance status, years with cancer, number of chronic disease, smoking habit, alcohol use, physical 52 53 activity, treatment regimen, district and BMI. Logistic regression models were used to measure the independent relationship 54 55 between HL and each of the QOL scale, while adjusting for other potentially confounding survivor characteristics. 56 57 The IRT analyses were conducted using the MULTILOG version 7.03. Other statistical analyses were performed with 58 59 SAS version 9.4 for each analysis, the null hypothesis was evaluated at a 2-sided significance level of 0.05. 60 5
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 21
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Patient and public involvement 5 6 Patients and public were not involved in the development of the research question or in the design of the study. 7 8 9 10 Results 11 12 Among 4713 cancer survivors surveyed, 4610 responded, for a 97.81% response rate. For 4589 of the 4610 participants, 13 14 at least 1 HL question was available in the survey; these participants composed our study sample. 15 16 The summed HL scale was measured by using all 3 HL questions. The correlations of single items to the total were 0.73, 17 18 0.70, and 0.75 for questionsFor 1 to 3, respectively. peer Figure review 1 shows the IRT modelsonly and test information for the total sample 19 20 and illustrates the implications of the item parameters for the response probabilities for different levels of health literacy. 21 22 All items had high discrimination parameters, except for Item 3, ‘How often do you have problems learning about your 23 24 medical condition because of difficulty reading hospital materials’, which had moderate discrimination (a=0.88) but still 25 26 satisfy the condition of a>0.75. The difficulty parameters (Threshold) increased monotonically across rating scale 27 28 categories for all items, ranging from -1.81 to 3.94, indicating that the 3 HL questions discriminate well. Among the 4589 29 30 participants in the study, the mean HL score was 6.3 (range, 3-15). As table 1 shows, one hundred fifty-nine participants 31 32 (3.5%) had low HL (HL score, 11-15). Participants with low HL were more likely than participants with adequate HL to 33 34 be older, to have received only some middle school education or less, and to have had a little exercise. 35 36 Table 2 presents the bivariate relationships between predictors of QOL and cancer survivors’ global health status. The
37 http://bmjopen.bmj.com/ 38 HL score, age, sex, marriage, years with cancer, number of chronic diseases, smoking habit, alcohol use, physical activity, 39 40 district, and BMI were all associated with QL. After adjustment for other potentially confounding factors, marriage, years 41 42 with cancer, and smoking habit were no significant differences. For each 1-point decrement in the HL score, the QL value 43 44 increased by 2.07 (P<.001).
45 on September 23, 2021 by guest. Protected copyright. 46 Cancer survivors with adequate HL were more likely than cancer survivors with low HL to have a high score for a 47 48 functional scale whereas a low level of symptomatology. According to the recommendation(24), a difference on the 0–100 49 50 51 scale of more than 5 points were considered as a clinically important difference. As can be seen from the data in Figure 2 52 53 that there had a clinically significant difference between survivors with adequate HL and survivors with low HL, in terms 54 55 of global health status (difference of adequate HL vs low HL [diff], 9.05), role function (diff: 5.2), emotional function (diff: 56 57 5.17), cognitive function (diff: 7.13), social function (diff: 7.49), insomnia (diff: -8.8), and financial difficulties (diff: - 58 59 8.71). 60 6
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 21 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Eighty-three percent of cancer survivors with adequate HL and 65% of cancer survivors with low HL reported that they 5 6 had a good global health status (unadjusted odds ratio [OR], 2.75; 95% confidence interval [CI], 1.97-3.84; P<.001). After 7 8 confounders were adjusted, cancer survivors with adequate HL were more likely to report a better global health status 9 10 (adjusted OR, 2.81; 95% CI, 1.94-4.06; P<.001) (Table 3). The extent of the associations between HL and other scales of 11 12 QOL, including four functional scales (physical, emotional, cognitive and social functioning), three symptom scales 13 14 (fatigue, pain, and nausea/vomiting), and three single-item (insomnia, appetite loss, and financial difficulties), was similar 15 16 to that of global health status and all had a statistical significance (Table 3). 17 18 Discussion For peer review only 19 20 Our study demonstrates that inadequate HL as assessed by 3 brief screening questions was present in less than 1 in 25 21 22 cancer survivors in Shanghai Cancer Rehabilitation Club. And inadequate HL was an independent predictor of poor QOL 23 24 and was associated with a lower level of functioning and a higher level of symptomatology or problems. The association 25 26 between HL and QOL that we observed is significant from a clinical and public health perspective. These findings highlight 27 28 a potential target for interventions to improve the overall quality of life for cancer survivors. 29 30 To our knowledge, few studies had demonstrated the association between HL and QOL among cancer survivors using 31 32 the 3 brief screening questions. Prior research studies have used more complex and extensive questionnaires to measure 33 34 HL, such as the Rapid Estimate of Adult Literacy in Medicine (25, 26), Cancer Health Literacy Test-30 (CHLT-30) (27), 35 36 the Short-Form Test of Functional Health Literacy in Adults (S-TOFHLA) (5, 28, 29), et al. which are also impractical for
37 http://bmjopen.bmj.com/ 38 use in busy clinical settings. Despite differences in the nature of HL assessment techniques, our results are consistent with 39 40 those of previous studies that identified inadequate HL as a risk factor for QOL among cancer survivors. (26, 14, 12) Our 41 42 results showing a higher adequate HL level are in contrast to those from other studies(30, 31) which found a lower adequate 43 44 HL in the Chinese population. This might because the participants we surveyed were all come from the Cancer
45 on September 23, 2021 by guest. Protected copyright. 46 Rehabilitation Club in Shanghai. The club (32) firmly “anti-cancer groups, beyond life”, and organizing a variety of 47 48 rehabilitation activities to make patients feel full confidence in their rehabilitation. They spread new concept of treatment 49 50 51 and health education mode their own health, such as “group psychotherapy”, “create a new life”, “qigong physical exercise”, 52 53 “five-full philosophy”, “participation in management”, et al. Through this process, participants in the club will have their 54 55 HL level gradually improved. This might be the reason why our results were higher than in other studies. 56 57 We chose QOL as the primary outcome of interest were because it can fully reflect the feeling and physical recovery 58 59 condition of the patients and has been increasingly used as a comprehensive health indicator in clinical treatment and 60 7
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 21
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 interventions (11, 33). QOL can be used as one of the indicators of efficacy of different therapeutic measures. The 5 6 combination of QOL and clinical curative effect observation index can help to choose the treatment that is more suitable 7 8 for patients. To evaluate the difference of QOL before and after treatment can help doctors have a whole understanding of 9 10 the patients’ true feelings and health so that they can in different stages of different measures. Studies have demonstrated 11 12 that there is a negative relationship between QOL and HL(26, 12), and inadequate health literacy is independently 13 14 associated with poor QOL (34). Consistent with this researches, our study showed that cancer survivors with adequate HL 15 16 had nearly 3 times the odds of having a better QOL than cancer survivors with inadequate HL. 17 18 From the public health perspective,For peer HL is an important review factor in ensuring only significant health outcomes. (12, 35, 36) 19 20 Inadequate HL may contribute to the disproportionate burden of cancer-related problems among disadvantaged populations. 21 22 The United Nations ECOSOC Ministerial Declaration of 2009 provided a clear mandate for action “call for the 23 24 development of appropriate action plans to promote health literacy.” (2) The 9th Global Conference was held in Shanghai, 25 26 China (2016) and addressed the determinants of health through good governance, health cities, health literacy, and social 27 28 mobilization.(2) Therefore, improving cancer survivors' health literacy is an important target for improving their QOL. 29 30 Besides that, participants with low HL were more likely than those with adequate HL to have received only some middle 31 32 school education or less and to have had a little exercise. Meanwhile, the education background and behaviors (alcohol use 33 34 and physical activity) were all associated with QL. Information needs and understand of the person living with a chronic 35 36 condition is critical for the optimal management (13). People with higher education levels have a better chance to acquire
37 http://bmjopen.bmj.com/ 38 health information about this kind of disease as well as understand doctors’ advice, and adopt a healthy lifestyle, such as 39 40 drink less alcohol and exercise regularly. The previous study reported that physical activity contributes to not only physical 41 42 health but to the emotional, social, cognitive and even the spiritual domain (37). This fit particularly with self-determination 43 44 theory that is when participants find physical activity could meet needs and contribute to QOL, move up the continuum
45 on September 23, 2021 by guest. Protected copyright. 46 toward more self-determined motivation. That positive cycle, with PA enhancing QOL, and enhanced QOL motivating 47 48 participation creates a positive health cycle (38). Meanwhile, recent cohorts study conducted in Hong Kong reported that 49 50
51 alcohol reduction to be associated with better mental well-being (39), which may have a positive effect on QOL. 52 53 Limitation 54 55 Several limitations should be considered in the interpretation of our findings. First, it was a cross-sectional study and 56 57 didn’t allow us to make a causal inference whether low HL was causally associated with poor QOL, and interventions 58 59 should be conducted in the further research. Second, we didn’t use the questionnaire of Chinese residents health literacy 60 8
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 21 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 monitoring and other complex tool to evaluate the HL, and the 3 brief questions may reflect other constructs, but the 3 brief 5 6 questions can easily be incorporated into the clinical routine work and are useful to identify high-risk patients. Third, this 7 8 study might only have a good representativeness of the cancer survivors in Shanghai Cancer Rehabilitation Club and may 9 10 not be generalizable to non-members of Cancer Rehabilitation Club. However, as discussed above, if the effect of 11 12 inadequate HL would be weaken in this participants, it may be greater in different populations. 13 14 Conclusions 15 16 In summary, inadequate health literacy as evaluated by 3 brief screening questions is independently associated with poor 17 18 QOL in cancer survivors. EffortsFor should peer focus on developing review and evaluating only interventions to improve QOL among cancer 19 20 survivors with inadequate HL. 21 22 DECLARATIONS 23 24 Ethics approval and consent to participate: Informed consent was obtained from all participants before enrollment. The 25 26 protocol was approved by the committee of Public Health School of Fudan University (protocol number IRB # 2017-05- 27 28 0621). 29 30 Consent for publication: Not applicable 31 32 Availability of data and material: Please contact author for data requests. 33 34 Competing interests: The authors declare that they have no competing interests. 35 36 Funding: Not applicable
37 http://bmjopen.bmj.com/ 38 Authors' contributions: Juan Xia carried out the field investigation, participated in the design of the study, performed the 39 40 statistical analysis and drafted the manuscript. Peng Wu in charge of the process of submission and revision of this paper. 41 42 Peng Wu, Qing-Long Deng, Rui Yan, Ren-Ren Yang and Bing-hui Lv participated in the field investigation and data 43 44 collection. Ji-Wei Wang and Jin-Ming Yu designed and coordinated the study and revised the manuscript. All authors read
45 on September 23, 2021 by guest. Protected copyright. 46 and approved the final manuscript. 47 48 Acknowledgements: We wish to thank Furong Tang, Donghui Yu, Mengxue Yin and some other volunteers who 49 50 51 contributed to this survey. We also wish to thank the thousands of cancer survivors who were willing to complete the 52 53 survey. 54 55 56 57 58 59 60 9
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 21
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Reference 5 6 1. Kindig, D. A., Panzer, A. M., & Nielsen-Bohlman, L. editors. Health Literacy: A Prescription to End Confusion. 7 8 National Academies Press 15, 2004:389-395. 9 10 2. World Health Organization. http://www.who.int/healthpromotion/conferences/9gchp/health-literacy/en/ 11 12 Accessed Februa Eichler K ry 1, 2018. 13 14 3. Eichler K, Wieser S, Brügger U. The costs of limited health literacy: a systematic review. International Journal 15 16 of Public Health. 2009; 54, 313. 17 18 4. Baker DW, Parker ForRM, Williams peer MV, et al. Thereview relationship of patient only reading ability to self-reported health and 19 20 use of health services. American Journal of Public Health..1997; 87, 1027. 21 22 5. Schillinger D, Grumbach K, Piette J, et al. Association of Health Literacy With Diabetes Outcomes. JAMA. 23 24 2002;288(4):475–482. 25 26 6. Weiss BD, Hart G, Mcgee DL, et al. Health status of illiterate adults: relation between literacy and health status 27 28 among persons with low literacy skills. Journal of the American Board of Family Practice.1992; 5, 257. 29 30 7. International Agency for Research on Cancer. GLOBOCAN 2012: Estimated Cancer Incidence, Mortality and 31 32 Prevalence Worldwide in 2012. http://globocan.iarc.fr/ Accessed February 1, 2018. 33 34 8. GRAYSON M. Breast cancer[J]. Nature, 2012,485(7400):S49. 35 36 9. KEYGHOBADI N, RAFIEMANESH H, MOHAMMADIAN-HAFSHEJANI A, et al. Epidemiology and trend
37 http://bmjopen.bmj.com/ 38 of cancers in the province of Kerman: southeast of Iran[J]. Asian Pac J Cancer Prev, 2015,16(4):1409-1413. 39 40 10. GHISLAIN I, ZIKOS E, COENS C, et al. Health-related quality of life in locally advanced and metastatic breast 41 42 cancer: methodological and clinical issues in randomised controlled trials[J]. Lancet Oncol, 2016,17(7):e294- 43 44 e304.
45 on September 23, 2021 by guest. Protected copyright. 46 11. Liu Z, Yang J. Health related quality of life: is it another comprehensive evaluation indicator of Chinese medicine 47 48 on acquired immune deficiency syndrome treatment? Journal of Traditional Chinese Medicine. 2015; 35(5):600-5. 49 50 51 12. Halverson JL, Martinez-Donate AP, Palta M, et al. Health Literacy and Health-Related Quality of Life Among a 52 53 Population-Based Sample of Cancer Patients. J Health Commun. 2015;20(11):1320-9. 54 55 13. KUGBEY N, MEYER-WEITZ A, OPPONG A K. Access to health information, health literacy and health-related 56 57 quality of life among women living with breast cancer: Depression and anxiety as mediators[J]. Patient Educ 58 59 Couns, 2019. 60 10
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 21 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 14. Husson O, Mols F, Fransen MP, et al. Low subjective health literacy is associated with adverse health behaviors 5 6 and worse health-related quality of life among colorectal cancer survivors: results from the profiles registry. 7 8 Psychooncology. 2015;24(4):478-86. 9 10 15. HAHN E A, CELLA D, DOBREZ D G, et al. The impact of literacy on health-related quality of life measurement 11 12 and outcomes in cancer outpatients[J]. Qual Life Res, 2007,16(3):495-507. 13 14 16. Peterson PN, Shetterly SM, Clarke CL,et al. Health Literacy and Outcomes Among Patients With Heart Failure. 15 16 JAMA. 2011;305(16):1695–1701. 17 18 17. Chew LD, Griffin JM,For Partin MR,peer et al. Validation review of screening questions only for limited health literacy in a large VA 19 20 outpatient population. J Gen Intern Med. 2008; 23(5), 561-566. 21 22 18. Wallace LS, Rogers ES, Roskos SE, et al. Brief report: screening items to identify patients with limited health 23 24 literacy skills. J Gen Intern Med. 2006;21(8):874-7. 25 26 19. Wallace LS, Cassada DC, Rogers ES, et al. Can screening items identify surgery patients at risk of limited health 27 28 literacy? J Surg Res. 2007;140(2):208-13. 29 30 20. Chew LD, Bradley KA, Boyko EJ. Brief questions to identify patients with inadequate health literacy. Fam Med. 31 32 2004;36(8):588-94. 33 34 21. Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization for Research and Treatment of Cancer 35 36 QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst.
37 http://bmjopen.bmj.com/ 38 1993;85(5):365-76. 39 40 22. Wang JW. Tang Z, Yu JM, et al. QLQ-BR23 and EORTC+QLQ-C30 for measurement of the impact of 41 42 rehabilitation exercise on quality of life in breast cancer patients. Fudan Univ J Med Sci, 2015. 43 44 23. Fayers PM, Aaronson NK, Bjordal K. The EORTC QLQ-C30 scoring manual. 2001.
45 on September 23, 2021 by guest. Protected copyright. 46 24. Osoba D, Rodrigues G, Myles J. Interpreting the significance of changes in health-related quality-of-life scores. 47 48 J Clin Oncol. 1998;16(1):139-44. 49 50 51 25. Davis TC, Long SW, Jackson RH, et al. Rapid estimate of adult literacy in medicine: a shortened screening 52 53 instrument. Fam Med. 1993;25(6):391-5. 54 55 26. Song L, Mishel M, Bensen JT, et al. How does health literacy affect quality of life among men with newly 56 57 diagnosed clinically localized prostate cancer? Findings from the North Carolina-Louisiana Prostate Cancer 58 59 Project (PCaP). Cancer. 2012;118(15):3842-51. 60 11
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 21
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 27. Echeverri M, Anderson D, Nápoles AM. Cancer Health Literacy Test-30-Spanish (CHLT-30-DKspa), a New 5 6 Spanish-Language Version of the Cancer Health Literacy Test (CHLT-30) for Spanish-Speaking Latinos. J Health 7 8 Commun. 2016;21 Suppl 1:69-78. 9 10 28. Yehle KS, Plake KS, Nguyen P, et al. Health-Related Quality of Life in Heart Failure Patients With Varying 11 12 Levels of Health Literacy Receiving Telemedicine and Standardized Education. Home Healthc Now. 13 14 2016;34(5):267-72. 15 16 29. Baker DW, Williams MV, Parker RM, et al. Development of a brief test to measure functional health literacy. 17 18 Patient Educ Couns.For 1999;38(1):33-42. peer review only 19 20 30. Yao H, Shi Q, Li Y. The Current Status of Health Literacy in China. China Population Today. 2016; 41-41. 21 22 31. Huang L, Peng S, University FM, et al. The Status and Influencing Factors of Health Literacy among Cancer 23 24 Patients. Nursing Journal of Chinese Peoples Liberation Army. 2016. \ 25 26 32. http://www.shcrc.cn/webs/intro.aspx. Published 2012. Accessed February 1, 2018. 27 28 33. Guyatt GH, Feeny DH, Patrick DL. Measuring Health-Related Quality-of-Life. Ann Intern Med. 1993; 29 30 118(8):622-9. 31 32 34. Wang CL, Kane RL, Xu DJ, et al. Health literacy as a moderator of health-related quality of life responses to 33 34 chronic disease among Chinese rural women. BMC Womens Health. 2015;15:34. 35 36 35. Protheroe J, Wolf MS, Lee A. Health literacy and health outcomes. Health Literacy in Context: International
37 http://bmjopen.bmj.com/ 38 Perspectives. 2010;57-74. 39 40 36. Aaby A, Friis K, Christensen B, et al. Health literacy is associated with health behaviour and self-reported health: 41 42 A large population-based study in individuals with cardiovascular disease. Eur J Prev Cardiol. 2017;24(17):1880- 43 44 1888.
45 on September 23, 2021 by guest. Protected copyright. 46 37. GILL D L, HAMMOND C C, REIFSTECK E J, et al. Physical Activity and Quality of Life[J]. Journal of 47 48 Preventive Medicine & Public Health, 2013,46(Suppl 1):S28-S34. 49 50 51 38. DECI E L, RYAN R M. The \"What\" and \"Why\" of Goal Pursuits: Human Needs and the Self-Determination 52 53 of Behavior[J]. Psychological Inquiry, 2000,11(4):227-268. 54 55 39. YAO X I, NI M Y, CHEUNG F, et al. Change in moderate alcohol consumption and quality of life: evidence 56 57 from 2 population-based cohorts[J]. CMAJ, 2019,191(27):E753-E760. 58 59 60 12
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 21 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Figure Title: 5 6 Figure 1. Item Characteristic Curve and Test Information Curve for the 3 established screening questions 7 8 Figure 2. Unadjusted QOL scores for cancer survivors with inadequate and adequate health literacy 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 13
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 21
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Table 1. Characteristics of Respondent stratified by Health Literacy Level* 5 6 Level of Health Literacy 7 8 Characteristics Total (N=4589) Adequate (n=4430) Inadequate (n=159) P value 9 10 Age, M (Q1,Q3), y 62.0 (57.04,66.53) 62.0 (57.00,66.53) 62.52 (58.37,68.39) 0.0485 11 12 Years with cancers, M (Q1,Q3) 6.09 (3.50,10.50) 6.09 (3.50,10.50) 5.92 (3.50,10.58) 0.6258 13 14 Sex, No.(%) 0.9424 15 16 Male 1057 (23.03) 1020 (96.50) 37 (3.50) 17 18 Female For peer3532 (76.97) review3410 (96.55) only122 (3.45) 19 20 Race 0.8644 21 22 Hanzu 4544 (99.28) 4386 (96.52) 158 (3.48) 23 24 Marriage 0.2192 25 26 single 129 (2.81) 121 (93.80) 8 (6.20) 27 28 married/ cohabitation 3989 (86.93) 3853 (96.61) 136 (3.41) 29 30 divorced/widowed/ separated 471 (10.226) 456 (96.82) 15 (3.18) 31 32 Annual income, ¥ 0.0534 33 34 <5000 4134 (91.89) 3987 (96.44) 147 (3.56) 35 36 ≥5000 365 (8.11) 359 (98.36) 6 (1.64)
37 http://bmjopen.bmj.com/ 38 Family income 0.8296 39 40 <5000 3086 (70.07) 2979 (96.53) 107 (3.47) 41 42 ≥5000 1318 (29.93) 1274 (96.66) 44 (3.34) 43 44 Education 0.0003
45 on September 23, 2021 by guest. Protected copyright. 46 47 Middle school or less 4021 (87.62) 3867 (96.17) 154 (3.83) 48 49 High school or college 568 (12.38) 563 (99.12) 5 (0.88) 50 51 Insurance status 0.8159 52 53 Uninsured 645 (15.55) 623 (96.59) 22 (3.41) 54 55 Medicare 3504 (84.45) 3378 (96.40) 126 (3.60) 56 57 Treatment regimen 0.2753 58 59 Chemotherapy 661 (14.44) 641 (96.97) 20 (3.03) 60 14
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 21 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Radiotherapy 74 (1.62) 68 (91.89) 6 (8.11) 5 6 Operation 931 (20.34) 902 (96.89) 29 (3.11) 7 8 Chemotherapy & Radiotherapy 162 (3.54) 155 (95.68) 7 (4.32) 9 10 Chemotherapy & Radiotherapy 897 (19.59) 870 (96.99) 27 (3.01) 11 12 & Operation 13 14 Chemotherapy & Operation 1658 (36.22) 1594 (96.14) 64 (3.86) 15 16 Biotherapy 69 (1.51) 66 (95.65) 3 (4.35) 17 18 Number of Chronic diseaseFor peer review only 0.3564 19 20 0 1238 (26.98) 1203 (97.17) 35 (2.83) 21 22 1-3 2735 (59.60) 2634 (96.31) 101 (3.69) 23 24 >3 616 (13.42) 593 (96.27) 23 (3.73) 25 26 Smoking habit 0.5284 27 28 Never 3841 (83.70) 3711 (96.62) 130 (3.38) 29 30 Former 598 (13.03) 573 (95.82) 25 (4.18) 31 32 Current 150 (3.27) 146 (97.33) 4 (2.67) 33 34 Current alcohol use 0.0637 35 36 None 4264 (92.92) 4113 (96.46) 151 (3.54)
37 http://bmjopen.bmj.com/ 38 Light to moderate 215 (4.69) 213 (99.07) 2 (0.93) 39 40 Heavy 110 (2.40) 104 (94.55) 6 (5.45) 41 42 Body mass index# 0.4593 43 44 <18.5 233 (5.16) 229 (98.28) 4 (1.72)
45 on September 23, 2021 by guest. Protected copyright. 46 18.5-23.9 2342 (51.85) 2258 (96.41) 84 (3.59) 47 48 24.0-27.9 1546 (34.23) 1492 (96.51) 54 (3.49) 49 50 51 ≥28.0 396 (8.77) 380 (95.96) 16 (4.04) 52 53 Physical activity 0.0264 54 55 No 1038 (22.64) 991 (95.47) 47 (4.53) 56 57 1-4 times per week 2493 (54.37) 2406 (96.51) 87 (3.49) 58 59 ≥5 times per week 1054 (22.99) 1029 (97.63) 25 (2.37) 60 15
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 21
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 * Data are given as percentages unless otherwise indicated. 5 6 # Calculated as weight in kilograms divided by the square of height in meters. 7 8 9 10 Table 2. Relationship Between Survivors Characteristics and QOL 11 12 Unadjusted Adjusted* 13 Predictor 14 Coefficientτ P Value Coefficientτ P Value 15 16 HL score -2.11 <0.0001 -2.07 <0.0001 17 18 Age For -0.25peer review<0.0001 only-0.16 0.0053 19 20 Sex, female -3.92 <0.0001 -5.02 0.0002 21 22 Race/ethnicity 2.10 0.5854 6.41 0.1819 23 24 Marriage -3.01 0.0051 -2.14 0.0710 25 26 Education -0.28 0.6316 -1.38 0.0319 27 28 Annual income 0.42 0.7664 0.81 0.6044 29 30 Family income -0.03 0.9694 0.53 0.5521 31 32 Insurance status -1.67 0.1263 -0.09 0.9346 33 34 Years with cancer (per year) -0.23 0.0002 -0.04 0.5401 35 36 Number of Chronic disease -5.79 <0.0001 -4.87 <0.0001
37 http://bmjopen.bmj.com/ 38 Smoking habit 1.35 0.015 -1.23 0.1210 39 40 Alcohol intake 3.38 <0.0001 2.07 0.0240 41 42 Physical activity 4.07 <0.0001 3.15 <0.0001 43 44 Treatment regimen -0.004 0.985 0.16 0.4474
45 on September 23, 2021 by guest. Protected copyright. 46 47 District 0.63 <0.0001 0.48 <0.0001 48 49 BMI 1.14 0.0305 2.18 <0.0001 50 51 * Adjusted for age, sex, race, marriage, income, education, insurance, years with cancer, treatment regimen, district, 52 53 number of chronic disease, smoking habit, alcohol use, physical activity and BMI. 54 55 τAll coefficients correspond to a change in QOL score for unit change of each covariate. 56 57 58 59 60 16
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 21 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Table 3. Adjusted Odds Ratio of QOL for cancer survivors with inadequate vs adequate health literacy* 5 6 QOL Survivors with poor QOL Odds Ratio (95%CI) P value 7 8 Global health status (QL) 788 2.81 (1.94-4.06) <0.0001 9 10 Physical function (PF) 83 2.84 (1.18-6.87) 0.0205 11 12 Role function (RF) 53 2.70 (0.94-7.80) 0.0664 13 14 Emotional function (EF) 77 5.01 (2.37-10.58) <0.0001 15 16 Cognitive function (CF) 128 4.07 (2.18-7.61) <0.0001 17 18 Social function (SF) For 250peer review3.59 only (2.19-5.88) <0.0001 19 20 Fatigue (FA) 425 2.63 (1.68-4.12) <0.0001 21 22 Nausea/Vomiting (NV) 52 5.39 (2.17-13.38) 0.0003 23 24 Pain (PA) 296 2.37 (1.41-3.97) 0.0011 25 26 Dyspnea (DY) 152 1.86 (0.88-3.94) 0.106 27 28 Insomnia (SL) 475 2.41 (1.56-3.74) <0.0001 29 30 Appetite loss (AP) 91 3.84(1.84-8.03) 0.0003 31 32 Constipation (CO) 188 1.06 (0.45-2.46) 0.8995 33 34 Diarrhea (DI) 114 2.11 (0.95-4.71) 0.0679 35 36 Financial difficulties (FI) 747 2.73 (1.88-3.97) <0.0001
37 http://bmjopen.bmj.com/ 38 *Adjusted for age, sex, race, marriage, income, education, insurance, years with cancer, treatment regimen, district, number 39 40 of chronic disease, smoking habit, alcohol use, physical activity and BMI 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 21
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 Item 1 Item 2 6 Slope=1.49 Slope= 2.21 7 Threshold1= -1.03 Threshold1= -0.15 Threshold2= 0.81 Threshold2= 1.21 8 Threshold3= 2.48 Threshold3= 2.02 9 Threshold4= 3.80 Threshold4= 2.48 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 Item 3 24 Slope= 0.88 25 Threshold1= -1.81 Threshold2= -0.05 26 Threshold3= 2.47 27 Threshold4= 3.94 28 29 30 31 32 33 34 35 36
37 http://bmjopen.bmj.com/ 38 39 Figure 1. Item Characteristic Curve and Test Information Curve for the 3 established screening questions 40 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 21 BMJ Open
1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 2. Unadjusted QOL scores for cancer survivors with inadequate and adequate health literacy 33 34 35 36
37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 21
1 2 STROBE Statement—Checklist of items that should be included in reports of cross-sectional studies
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Item 5 No Recommendation 6 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or 7 (Page 2, Line 6) the abstract 8 (Page 2, Line 25) (b) Provide in the abstract an informative and balanced summary of what 9 10 was done and what was found 11 Introduction 12 Background/rationale 2 Explain the scientific background and rationale for the investigation being 13 14 reported 15 Objectives (Page 3, Line 27-28) 3 State specific objectives, including any prespecified hypotheses 16 Methods 17 18 Study design (Page 4, LineFor 3) peer4 Present keyreview elements of study only design early in the paper 19 Setting (Page 4, Line 3-7) 5 Describe the setting, locations, and relevant dates, including periods of 20 recruitment, exposure, follow-up, and data collection 21 22 Participants (Page 4, Line 8-18) 6 (a) Give the eligibility criteria, and the sources and methods of selection of 23 participants 24 Variables (Page 4, Line 18-Page 7 Clearly define all outcomes, exposures, predictors, potential confounders, 25 5, Line 14) and effect modifiers. Give diagnostic criteria, if applicable 26 27 Data sources/ measurement 8* For each variable of interest, give sources of data and details of methods 28 (Page 4, Line 18-Page 5, Line 7) of assessment (measurement). Describe comparability of assessment 29 methods if there is more than one group 30 Bias (Page 8, Line 24) 9 Describe any efforts to address potential sources of bias 31 32 Study size 10 Explain how the study size was arrived at 33 (Page 4, Line 16-18) 34 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If 35 (Page 5, Line 21-22) applicable, describe which groupings were chosen and why 36 Statistical methods 12 (a) Describe all statistical methods, including those used to control for
37 http://bmjopen.bmj.com/ 38 (Page 5, Line 16-29) confounding 39 (b) Describe any methods used to examine subgroups and interactions 40 41 (c) Explain how missing data were addressed 42 (d) If applicable, describe analytical methods taking account of sampling 43 strategy 44 (e) Describe any sensitivity analyses
45 on September 23, 2021 by guest. Protected copyright. 46 Results 47 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers 48 (Page 6, Line 5-6) potentially eligible, examined for eligibility, confirmed eligible, included 49 50 in the study, completing follow-up, and analysed 51 (b) Give reasons for non-participation at each stage 52 (c) Consider use of a flow diagram 53 54 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, 55 (Page 6, Line 7-16) social) and information on exposures and potential confounders 56 (b) Indicate number of participants with missing data for each variable of 57 interest 58 59 Outcome data 15* Report numbers of outcome events or summary measures 60 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted (Page 6, Line 17- Page 7, Line 6) estimates and their precision (eg, 95% confidence interval). Make clear
1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 21 BMJ Open
1 2 which confounders were adjusted for and why they were included
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from (b) Report category boundaries when continuous variables were 4 5 categorized 6 (c) If relevant, consider translating estimates of relative risk into absolute 7 risk for a meaningful time period 8 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, 9 10 and sensitivity analyses 11 Discussion 12 Key results 18 Summarise key results with reference to study objectives 13 14 (Page 7, Line 8-12) 15 Limitations 19 Discuss limitations of the study, taking into account sources of potential 16 (Page 8, Line 24-Page 9, l, Line bias or imprecision. Discuss both direction and magnitude of any potential 17 18 4) For peerbias review only 19 Interpretation 20 Give a cautious overall interpretation of results considering objectives, 20 (Page 7, Line 13-Page 8, Line limitations, multiplicity of analyses, results from similar studies, and other 21 24) relevant evidence 22 23 Generalisability 21 Discuss the generalisability (external validity) of the study results 24 (Page 9, Line 5) 25 Other information 26 27 Funding 22 Give the source of funding and the role of the funders for the present study 28 (Page 9, Line 16) and, if applicable, for the original study on which the present article is 29 based 30 31 32 *Give information separately for exposed and unexposed groups. 33 34 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 35 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 36 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at
37 http://bmjopen.bmj.com/ 38 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 39 available at www.strobe-statement.org. 40 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from
Relationship between health literacy and quality of life among cancer survivors in China: a cross-sectional study
Journal: BMJ Open ManuscriptFor ID peerbmjopen-2018-028458.R2 review only Article Type: Original research
Date Submitted by the 18-Sep-2019 Author:
Complete List of Authors: Xia, Juan; Fudan University, school of public health Wu, Peng; Fudan University, school of public health Deng, Qinglong; Fudan University, school of public health Yan, Rui; Fudan University, school of public health Yang, Renren; Fudan University, school of public health Lv, Binghui; Fudan University, school of public health Wang, Jiwei; Fudan University, Fudan university school of public health Yu, Jinming; Fudan University, school of public health
Primary Subject Public health Heading:
Secondary Subject Heading: Oncology
Keywords: Health Literacy, 3 Brief Questions, Quality of Life, Cancer Survivors http://bmjopen.bmj.com/
on September 23, 2021 by guest. Protected copyright.
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 23 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Title Page 5 6 Title: Relationship between health literacy and quality of life among cancer survivors in 7 8 China: a cross-sectional study 9 10 11 12 Authors full name, highest academic degrees and institutional affiliations: 13 14 Juan Xia1,2, PhD, Peng Wu2, Qing-Long Deng2, Rui Yan2, Ren-Ren Yang2, Bing-hui Lv2, Ji- 15 16 Wei Wang2, PhD, Jin-Ming Yu2, PhD 17 18 1, Department of ForEpidemiology peer and Health review Statistics, School only of Public Health, Capital Medical 19 20 University, Beijing, China. Beijing Municipal Key Laboratory of Clinical Epidemiology, 21 22 Capital Medical University, Beijing, China 23 24 2, Institute of Clinical Epidemiology, Key Laboratory of Public Health Safety, Ministry of 25 26 Education, Key Lab of Health Technology Assessment of Ministry of Health, School of Public 27 28 Health, Fudan University 29 30 Juan Xia and Peng Wu have contributed equally to this work. 31 32 33 34 Information for Corresponding Author: 35 36 Jin-Ming Yu, School of Public Health, Fudan University, 130 Dong-An Road, Shanghai, China,
37 http://bmjopen.bmj.com/ 38 200032. Tel: (021) 54237447; Fax: (021)54237868; E-mail: [email protected]. 39 40 Ji-Wei Wang, School of Public Health, Fudan University, 130 Dong-An Road, Shanghai, China, 41 42 200032. Tel: (021)54237898; Fax: (021)54237898; E-mail: [email protected]. 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 Manuscript word count: 4805 words 47 48 49 Numbers of references: 45 50 51 Numbers of tables and figures: 3 tables and 2 figure 52 53 Abstract word count: 292 words 54 55 Key words: Health Literacy; 3 Brief Questions; Quality of Life; Cancer Survivors 56 57 Conflict of interest statement: None declared 58 59 60 1
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 23
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Relationship between health literacy and quality of life among cancer survivors in China: a cross- 5 6 sectional study 7 8 Objective To evaluate the association between Health Literacy (HL) and Quality of Life (QOL) among Cancer 9 10 Survivors in China. 11 12 Design Cross-sectional study in China. 13 14 Setting and participants Cross-sectional observational study of 4589 cancer surM vivors who were older than 18 15 16 years came from the Shanghai Cancer Rehabilitation Club. Participants were enrolled and completed questionnaires 17 18 between May and July 2017.For peer review only 19 20 Measurement HL were accessed by 3 established screening questions and QOL was evaluated using the simplified 21 22 Chinese version of the quality-of-life questionnaire-core 30 items (EORTC QLQ-30). All questionnaires were 23 24 collected through face-to-face interviews or self-administered by literate participants. Participants were excluded if 25 26 they did not complete at least 1 HL question. Baseline characteristics were compared by levels of HL using 휒2 test 27 28 for categorical variables and Wilcoxon rank sum test for the non-normal continuous variables. Item Response Theory 29 30 (IRT) was used to evaluate the existing measure of health literacy. Linear regression models and Logistic regression 31 32 models were used to investigate the association between HL and QOL. SAS 9.4 and MULTILOG 7.03 were applied 33 34 for analysis. 35 36 Results Valid selection of subjects in this study were 4589. IRT scaling parameters of the health literacy measure
37 http://bmjopen.bmj.com/ 38 found items had a good range of discriminating and difficulty. Of 4589 included responders, 159 (3.5%) had low HL. 39 40 After adjusting for participants’ sociodemographic characteristics, treatment regimen, and years with cancers, for 41 42 each 1-point decrement in HL score, the QOL score increased by 2.07 (P<0.001). Cancer survivors with low HL were 43 44 less likely than those with adequate HL to achieve a better QOL. In logistic regression, low HL was independently
45 on September 23, 2021 by guest. Protected copyright. 46 associated with poor QOL (adjusted odds ratio, 2.81 [95% confidence interval], 1.94-4.06; P<0.001). 47 48 Conclusions Among cancer survivors in Shanghai Cancer Rehabilitation Club, low HL was independently associated 49 50 51 with poor QOL. 52 53 Key words: Health Literacy; 3 Brief Questions; Quality of Life; Cancer Survivors 54 55 56 57 Strengths and limitations of this study 58 59 Few studies using 3 brief screening questions to evaluate Chinese cancer survivors’ health literacy instead of 60 2
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 23 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 the complex and long questions which are not suitable to use in clinical routine practice, and to explore its 5 6 relationship with quality of life. 7 8 Item response theory was used to evaluate whether a summed unidimensional health literacy scale could be 9 10 used. 11 12 Causal inferences could not be drawn due to the cross-sectional design employed. 13 14 The proportion of high HL might be overestimated, and low literacy patients are often excluded from the study 15 16 because of illiteracy. 17 18 For peer review only 19 Introduction 20 21 Health literacy (HL) is an evolving concept. As defined by the National Library of Medicine (1), HL is “the degree 22 23 to which individuals have the capacity to obtain, process, and understand basic health information and services 24 25 needed to make appropriate health decisions.” It means more than simply the ability to “read pamphlets”, “make 26 27 appointments”, “understand food labels” or “comply with prescribed actions” from a doctor (2). Higher levels of HL 28 29 within populations yield social benefits (3, 4), and HL is an important factor in ensuring significant health outcomes 30 31 32 (5, 6). Previous studies indicating that poor or limited HL can make it difficult for patients to function effectively in 33 34 the health care system, and is associated with several negative outcomes such as poor health status, decreased 35 36 comprehension of medical information, lack of engagement with doctors, higher cost of health care, and so forth (7,
37 http://bmjopen.bmj.com/ 38 8). 39 40 Cancer is one of the leading causes of morbidity and mortality worldwide (9) and the economic impact of cancer 41 42 is increasing. Despite the fact that China had a lower incidence rate of cancer compared with western counties, it 43 44 increased with a sharp slope in decades (10, 11). Meanwhile, patients nowadays are detected with advanced
45 on September 23, 2021 by guest. Protected copyright. 46 diagnostic techniques and cured by modern therapeutic methods, which resulted in the extension of lifespan of cancer 47 48 patients and more attention paid to the quality of life in their survival years (12). 49 50 Health-related quality of life (HRQOL) refers to individuals' subjective assessment of well-being and ability to 51 52 perform social roles, has been accepted as a health indicator in medical surroundings, like clinical intervention, 53 54 treatment, and health survey (13). Studies evaluating the relationship between HL and HRQOL produced mixed 55 56 results among cancer survivors. A longitudinal, population-based study conducted in Netherlands between 2000 and 57 58 2009 (n=1643, response rate 83%) showed that low subjective HL was associated with worse QOL among colorectal 59 60 cancer survivors who registered by the Eindhoven Cancer Registry (14). A study by Elizabeth A . Hahn et al. 3
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 23
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 conducted among 420 cancer outpatients enrolled at five Chicago-area cancer centers indicated that low literacy is 5 6 not an independent risk factor for poorer HRQL among cancer outpatients (15). There were other various studies on 7 8 different types of population reported different results and few similar study are conducted in less developed country, 9 10 like China. 11 12 A growing body of research measured HL with complex and long questions that are not suitable to use in clinical 13 14 routine practice (16). Thus the study adopted Chew’s three brief health literacy screening (BHLS) with established 15 16 validity for evaluating subjective functional HL (17). The scale only has three questions which can be easy and handy 17 18 implemented in busy clinicalFor settings. peer QOL were chose review as the primary outcomeonly of interest because it can fully reflect 19 20 the feeling and physical recovery condition of the patients and has been increasingly used as a comprehensive health 21 22 indicator in clinical treatment and interventions (18, 19). QOL can be used as one of the indicators of efficacy of 23 24 different therapeutic measures. The combination of QOL and clinical curative effect observation index can help to 25 26 choose the treatment that is more suitable for patients. To evaluate the difference of QOL before and after treatment 27 28 can help doctors have a whole understanding of the patients’ true feelings and health so that they can in different 29 30 stages of different measures. The simplified Chinese version of the quality-of-life questionnaire-core 30 items 31 32 (EORTC QLQ-30) was widely used to assess quality of life among cancer patients, and well‐documented validity 33 34 and reliability in various populations (20, 21). 35 36 Therefore, by using a population-based survey, this study aims to evaluate the association between HL and QOL
37 http://bmjopen.bmj.com/ 38 among sample of cancer survivors (breast, colorectal, lung, stomach, thyroid, and so on) in Shanghai Cancer 39 40 Rehabilitation Club using 3 brief screening questions. We hypothesize that higher HL levels would be associated 41 42 with better physical and mental well-being. Isolating the independent contribution of HL toward cancer survivors’ 43 44 QOL would have important clinical and public health implications, and it help relieve the conflict between the
45 on September 23, 2021 by guest. Protected copyright. 46 complexity of cancer care and health deficits, and ultimately improve patients’ HRQOL (22). 47 48 49 50 51 METHODS 52 53 Design and setting 54 55 A cross-sectional study was conducted from May to July 2017. All participants were recruited from Shanghai Cancer 56 57 Rehabilitation Club, a nongovernmental self-help mutual aid organization that contains 20 members of the branch 58 59 offices, 175 community block group, and more than 13,000 members. The present study covered 16 districts in 60 4
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 23 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Shanghai (Huangpu, Pudong, Xuhui, Changning, Putuo, Hongkou, Yangpu, Minhang, Baoshan, Jiading, Jinshan, 5 6 Songjiang, Qingpu, Fengxian, Chongming, and Jingan). 7 8 Participants 9 10 All subjects of our study were cancer survivors who took part in SCRC. The inclusion criteria for study enrollment 11 12 were as shown below: 1) at least 18-year-old; 2) have a pathological diagnosis of cancer; 3) able to independently 13 14 participate in the cancer rehabilitation club; 4) willingness to provide written informed consent; 5) no cognitive 15 16 impairment or psychotic disorder. The information leaflets about the content and purpose of this study and written 17 18 informed consent forms wereFor obtained peer from patients review who met the inclusion only criteria ahead of the investigation, and a 19 20 box of eggs was offered for their participation as gifts. Investigators who were all students of Fudan University were 21 22 trained and field investigation was conducted. Questionnaires were collected through face-to-face interviews with 23 24 the help of well-trained field workers or self-administered by literate participants. 4713 cancer survivors were 25 26 surveyed, 4610 got responses. Except for the incomplete (All three questions are not answered as incomplete), we 27 28 totally acquired 4589 valid questionnaires. The protocol was approved by the committee of the Public Health School 29 30 of Fudan University (protocol number IRB # 2017-05-0621). 31 32 Measures 33 34 HL was assessed using 3 established screening questions and categorized as adequate or inadequate (23-26). The 35 36 items included: “How often do you have someone help you read hospital materials?”, “How confident are you filling
37 http://bmjopen.bmj.com/ 38 out forms by yourself”, and “How often do you have problems learning about your medical condition because of 39 40 difficulty reading hospital materials?”. Each question was scored by participants on a 5-point scale. HL was evaluated 41 42 as a continuous and dichotomous variable. Based on prior literature (16), scores were summed, and participants were 43 44 categorized as low HL if their total score was greater than 10 and adequate HL if it was 10 or lower.
45 on September 23, 2021 by guest. Protected copyright. 46 QOL was assessed using the simplified Chinese version of the quality-of-life questionnaire-core 30 items (EORTC 47 48 QLQ-30), which incorporates nine multi-item scales (27): five functional scales (physical, role, cognitive, emotional, 49 50 51 and social functioning), three symptom scales (fatigue, pain, and nausea/vomiting), and a global health status/QOL 52 53 scale; and six single-item scales (dyspnea, insomnia, appetite loss, constipation, diarrhea, and financial difficulties). 54 55 The psychometric properties of the questionnaire had been previously evaluated (28). Scale scores were calculated 56 57 by averaging items within scales and transforming average scores linearly. All of the scales range in scores from 0 58 59 to100. A high score for a functional scale represents a high/healthy level of functioning whereas a high score for a 60 5
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 23
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 symptom scale or item represents a high level of symptomatology or problems. For more details on the scoring 5 6 procedures see the EORTC QLQ-C30 Scoring Manual (29). Participants were classified as having better QOL if their 7 8 scores were higher than 50 and poor QOL if it was lower than 50. 9 10 Other variables 11 12 Covariates were age, sex, race/ethnicity, socioeconomic status, education level, marital status (married vs single 13 14 or divorced/widowed), insurance status, years with cancer, smoking habit, alcohol use, physical activity, treatment 15 16 regimen, district, BMI and history of coexisting illnesses. Because coexisting illnesses may affect survivors QOL, 17 18 we measured coexisting illnessesFor conditionspeer by askingreview cancer survivors only whether they had ever been told by a 19 20 physician that they had chronic diseases, including hypertension, hyperlipemia, hyperuricemia, diabetes, heart 21 22 disease, stroke, respiratory disease, digestive system disease, and skeleton system diseases. All covariates were 23 24 determined at the time of the survey. 25 26 Statistical Analysis 27 28 To determine whether a summed unidimensional health literacy scale could be used, item response theory (IRT) 29 30 was used to evaluate the item parameter (discrimination parameter, threshold parameter), item characteristic curve 31 32 (ICC) and item information curve (IIF). Since the item responses are classified into ordered polytomous categories, 33 34 the Graded Response Model (GRM) was used. For the participants answering only 1 or 2 of the 3 questions, the score 35 36 for those questions was multiplied by 3 and 1.5, respectively (16).
37 http://bmjopen.bmj.com/ 38 Baseline characteristics were compared across levels of HL using 휒2 test for categorical variables and the 39 40 Wilcoxon rank sum test for the non-normal continuous variables. Linear regression models were used to estimate the 41 42 association between HL score and QOL after controlling for differences in participants’ characteristics, including 43 44 age, sex, race/ethnicity, marriage, education, income, insurance status, years with cancer, number of chronic disease,
45 on September 23, 2021 by guest. Protected copyright. 46 smoking habit, alcohol use, physical activity, treatment regimen, district and BMI. Logistic regression models were 47 48 used to measure the independent relationship between HL and each of the QOL scale, while adjusting for other 49 50 51 potentially confounding survivor characteristics. 52 53 The IRT analyses were conducted using the MULTILOG version 7.03. Other statistical analyses were performed 54 55 with SAS version 9.4 for each analysis, the null hypothesis was evaluated at a 2-sided significance level of 0.05. 56 57 Patient and public involvement 58 59 Patients and public were not involved in the development of the research question or in the design of the study. 60 6
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 23 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 6 Results 7 8 Among 4713 cancer survivors surveyed, 4610 responded, for a 97.81% response rate. For 4589 of the 4610 9 10 participants, at least 1 HL question was available in the survey; these participants composed our study sample and 11 12 the valid rate was 99.5%. 13 14 The summed HL scale was measured by using all 3 HL questions. The correlations of single items to the total were 15 16 0.73, 0.70, and 0.75 for questions 1 to 3, respectively. Figure 1 shows the IRT models and test information for the 17 18 total sample and illustratesFor the implications peer of the itemreview parameters for theonly response probabilities for different levels 19 20 of health literacy. All items had high discrimination parameters, except for Item 3, ‘How often do you have problems 21 22 learning about your medical condition because of difficulty reading hospital materials’, which had moderate 23 24 discrimination (a=0.88) but still satisfy the condition of a>0.75. The difficulty parameters (Threshold) increased 25 26 monotonically across rating scale categories for all items, ranging from -1.81 to 3.94, indicating that the 3 HL 27 28 questions discriminate well. Among the 4589 participants in the study, the mean HL score was 6.3 (range, 3-15). As 29 30 table 1 shows, one hundred fifty-nine participants (3.5%) had low HL (HL score, 11-15). Participants with low HL 31 32 were more likely than participants with adequate HL to be older, to have received only some middle school education 33 34 or less, and to have had a little exercise. 35 36 Table 2 presents the bivariate relationships between predictors of QOL and cancer survivors’ global health status.
37 http://bmjopen.bmj.com/ 38 The HL score, age, sex, marriage, years with cancer, number of chronic diseases, smoking habit, alcohol use, physical 39 40 activity, district, and BMI were all associated with QL. After adjustment for other potentially confounding factors, 41 42 marriage, years with cancer, and smoking habit were no significant differences among levels of HL. For each 1-point 43 44 decrement in the HL score, the QL value increased by 2.07 (P<.001).
45 on September 23, 2021 by guest. Protected copyright. 46 Cancer survivors with adequate HL were more likely to receive higher socres on the functional scale and lower 47 48 scores on symptom compared to those with low HL. According to the recommendation (30), a difference on the 0– 49 50 51 100 scale of more than 5 points were considered as a clinically important difference. As can be seen from the data in 52 53 Figure 2 that there had a clinically significant difference between survivors with adequate HL and survivors with low 54 55 HL, in terms of global health status (difference of adequate HL vs low HL [diff], 9.05), role function (diff: 5.2), 56 57 emotional function (diff: 5.17), cognitive function (diff: 7.13), social function (diff: 7.49), insomnia (diff: -8.8), and 58 59 financial difficulties (diff: -8.71). 60 7
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 23
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Eighty-three percent of cancer survivors with adequate HL and 65% of cancer survivors with low HL reported that 5 6 they had a good global health status (unadjusted odds ratio [OR], 2.75; 95% confidence interval [CI], 1.97-3.84; 7 8 P<.001). After confounders were adjusted, cancer survivors with adequate HL were more likely to report a better 9 10 global health status (adjusted OR, 2.81; 95% CI, 1.94-4.06; P<.001) (Table 3). The extent of the associations between 11 12 HL and other scales of QOL, including four functional scales (physical, emotional, cognitive and social functioning), 13 14 three symptom scales (fatigue, pain, and nausea/vomiting), and three single-item (insomnia, appetite loss, and 15 16 financial difficulties), was similar to that of global health status and all had a statistical significance (Table 3). 17 18 For peer review only 19 20 Discussion 21 22 Our study demonstrates that inadequate HL as assessed by 3 brief screening questions was present in less than 1 23 24 in 25 cancer survivors in Shanghai Cancer Rehabilitation Club. And inadequate HL was an independent predictor of 25 26 poor QOL and was associated with a lower level of functioning and a higher level of symptomatology or problems. 27 28 The association between HL and QOL that we observed is significant from a clinical and public health perspective. 29 30 These findings highlight a potential target for interventions to improve the overall quality of life for cancer survivors. 31 32 To our knowledge, few studies had demonstrated the association between HL and QOL among cancer survivors 33 34 using the 3 brief screening questions in China. Prior research studies have used more complex and extensive 35 36 questionnaires to measure HL, such as the Rapid Estimate of Adult Literacy in Medicine (22, 31), Cancer Health
37 http://bmjopen.bmj.com/ 38 Literacy Test-30 (CHLT-30) (32), the Short-Form Test of Functional Health Literacy in Adults (S-TOFHLA) (5, 33, 39 40 34), et al. which are also impractical for use in busy clinical settings. Despite differences in the nature of HL 41 42 assessment techniques, our results are consistent with those of previous studies that identified inadequate HL as a 43 44 risk factor for QOL among cancer survivors (5, 14, 35). Our results showing a higher adequate HL level are in contrast
45 on September 23, 2021 by guest. Protected copyright. 46 to those from other Chinese studies (36, 37) which found a lower adequate HL in the Chinese population. This might 47 48 because the participants we surveyed were all come from the Cancer Rehabilitation Club in Shanghai. The club (38) 49 50 51 firmly advocates “anti-cancer groups, beyond life”, and organizing a variety of rehabilitation activities to make 52 53 patients feel full confidence in their rehabilitation. Participants in the club will have their HL level gradually improved. 54 55 This might be the reason why our results were higher than in other studies. 56 57 The result of the study showed that many factors were associated with QOL, including HL score, age, sex, number 58 59 of chronic disease, alcohol intake, physical activity, district, and BMI among cancer survivors, which almost in line 60 8
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 23 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 with previous studies (39). Continuous HL score was found to be associated with QOL. The higher of HL scores 5 6 (indicating limited HL), the lower of QOL (35). Gender was found to be a determinant of QOL and males were more 7 8 likely have poor QOL compared to females. Study showed that older survivors reported higher QOL, however, there 9 10 existed some research reported that overall QOL increased with age (40). There was strong association between QOL 11 12 and chronic disease, and QOL was lower in the presence of survivors with chronic diseases (41). Physical activity 13 14 was found to be positively associated with QOL. And those living in urban area were more likely to have better QOL 15 16 than those living in rural area. Survivors with higher BMI had higher overall QOL when controlling for confounders, 17 18 which was slightly differentFor from some peer studies conducted review among female only that reported that higher BMI was associated 19 20 with QOL (42, 43). 21 22 Besides that, the positive relationship between HL and scales of HRQOL were existed regardless of how HL or 23 24 HRQOL was operationalized (continuous or categorical). Our study showed that cancer survivors with adequate HL 25 26 had nearly 3 times the odds of having a better QOL than cancer survivors with inadequate HL and the same trend 27 28 could be found in other functional subscales. In symptom subscales, those with adequate HL were more likely to 29 30 have slight symptoms compared with those with limited HL. That’s largely because information needs and understand 31 32 of the person living with a chronic condition is critical for the optimal management (8). Kim et al. (44) indicated that 33 34 low health literacy prostate cancer patients may have hindered patient involvement in shared decision-making with 35 36 a physician. Those who had inadequate HL may have difficulty in understanding medical information given by
37 http://bmjopen.bmj.com/ 38 provider, managing their treatment plan, and adherence to cancer treatment regiments, resulting in exacerbated 39 40 treatment-related symptoms (35). Whereas, those who had adequate HL have better chance to acquire health 41 42 information about this kind of disease as well as understand doctors’ advice, and adopt a healthy lifestyle, such as 43 44 drink less alcohol and exercise regularly.
45 on September 23, 2021 by guest. Protected copyright. 46 From the public health perspective, HL is an important factor in ensuring significant health outcomes. (35, 45, 46) 47 48 Inadequate HL may contribute to the disproportionate burden of cancer-related problems among disadvantaged 49 50 51 populations. The United Nations ECOSOC Ministerial Declaration of 2009 provided a clear mandate for action “call 52 th 53 for the development of appropriate action plans to promote health literacy” (2). The 9 Global Conference was held 54 55 in Shanghai, China (2016) and addressed the determinants of health through good governance, health cities, health 56 57 literacy, and social mobilization (2). Therefore, improving cancer survivors' health literacy is an important target for 58 59 improving their QOL. 60 9
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 23
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 6 Limitation 7 8 Several limitations should be considered in the interpretation of our findings. First, it was a cross-sectional study 9 10 and causal inferences could not be drawn due to the cross-sectional design employed, and interventions should be 11 12 conducted in the further research. Second, we didn’t use the questionnaire of Chinese residents health literacy 13 14 monitoring and other complex tool to evaluate the HL, and the 3 brief questions may reflect other constructs, but the 15 16 3 brief questions can easily be incorporated into the clinical routine work and are useful to identify high-risk patients. 17 18 Third, this study might onlyFor have a good peer representativeness review of the cancer only survivors in Shanghai Cancer Rehabilitation 19 20 Club and may not be generalizable to non-members of Cancer Rehabilitation Club. However, as discussed above, if 21 22 the effect of inadequate HL would be weaken in this participants, it may be greater in different populations. 23 24 25 26 Conclusions 27 28 In summary, inadequate health literacy as evaluated by 3 brief screening questions is independently associated 29 30 with poor QOL in cancer survivors. Efforts should focus on developing and evaluating interventions to improve QOL 31 32 among cancer survivors with inadequate HL. 33 34 35 36 DECLARATIONS
37 http://bmjopen.bmj.com/ 38 Ethics approval and consent to participate: Informed consent was obtained from all participants before enrollment. 39 40 The protocol was approved by the committee of Public Health School of Fudan University (protocol number IRB # 41 42 2017-05-0621). 43 44 Consent for publication: Not applicable
45 on September 23, 2021 by guest. Protected copyright. 46 Availability of data and material: Please contact author for data requests. 47 48 Competing interests: None declared 49 50 Funding: Not applicable 51 52 Authors' contributions: Juan Xia carried out the field investigation, participated in the design of the study, 53 54 performed the statistical analysis and drafted the manuscript. Peng Wu in charge of the process of submission and 55 56 revision of this paper. Peng Wu, Qing-Long Deng, Rui Yan, Ren-Ren Yang and Bing-hui Lv participated in the field 57 58 investigation and data collection. Ji-Wei Wang and Jin-Ming Yu designed and coordinated the study and revised the 59 60 manuscript. All authors read and approved the final manuscript. 10
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 23 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Acknowledgements: We wish to thank Furong Tang, Donghui Yu, Mengxue Yin and some other volunteers who 5 6 contributed to this survey. We also wish to thank the thousands of cancer survivors who were willing to complete the 7 8 survey. 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 11
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 23
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Reference 5 6 1. Kindig, D. A., Panzer, A. M., & Nielsen-Bohlman, L. editors. Health Literacy: A Prescription to End 7 8 Confusion. National Academies Press 15, 2004:389-395. 9 10 2. World Health Organization. http://www.who.int/healthpromotion/conferences/9gchp/health-literacy/en/ 11 12 Accessed Februa Eichler K ry 1, 2018. 13 14 3. Eichler K, Wieser S, Brügger U. The costs of limited health literacy: a systematic review. International 15 16 Journal of Public Health. 2009; 54, 313. 17 18 4. Baker DW, ParkerFor RM, Williams peer MV, et al. review The relationship of only patient reading ability to self-reported health 19 20 and use of health services. American Journal of Public Health..1997; 87, 1027. 21 22 5. Schillinger D, Grumbach K, Piette J, et al. Association of Health Literacy With Diabetes Outcomes. JAMA. 23 24 2002;288(4):475–482. 25 26 6. Weiss BD, Hart G, Mcgee DL, et al. Health status of illiterate adults: relation between literacy and health 27 28 status among persons with low literacy skills. Journal of the American Board of Family Practice.1992; 5, 29 30 257. 31 32 7. JAYASINGHE U W, HARRIS M F, PARKER S M, et al. The impact of health literacy and life style risk 33 34 factors on health-related quality of life of Australian patients[J]. Health Qual Life Outcomes, 2016,14:68. 35 36 8. KUGBEY N, MEYER-WEITZ A, OPPONG A K. Access to health information, health literacy and health-
37 http://bmjopen.bmj.com/ 38 related quality of life among women living with breast cancer: Depression and anxiety as mediators[J]. 39 40 Patient Educ Couns, 2019. 41 42 9. International Agency for Research on Cancer. GLOBOCAN 2012: Estimated Cancer Incidence, Mortality 43 44 and Prevalence Worldwide in 2012. http://globocan.iarc.fr/ Accessed February 1, 2018.
45 on September 23, 2021 by guest. Protected copyright. 46 10. GRAYSON M. Breast cancer[J]. Nature, 2012,485(7400):S49. 47 48 11. KEYGHOBADI N, RAFIEMANESH H, MOHAMMADIAN-HAFSHEJANI A, et al. Epidemiology and 49 50 51 trend of cancers in the province of Kerman: southeast of Iran[J]. Asian Pac J Cancer Prev, 2015,16(4):1409- 52 53 1413. 54 55 12. GHISLAIN I, ZIKOS E, COENS C, et al. Health-related quality of life in locally advanced and metastatic 56 57 breast cancer: methodological and clinical issues in randomised controlled trials[J]. Lancet Oncol, 58 59 2016,17(7):e294-e304. 60 12
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 23 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 13. WANG H M, BEYER M, GENSICHEN J, et al. Health-related quality of life among general practice patients 5 6 with differing chronic diseases in Germany: cross sectional survey[J]. BMC Public Health, 2008,8:246. 7 8 14. Husson O, Mols F, Fransen MP, et al. Low subjective health literacy is associated with adverse health 9 10 behaviors and worse health-related quality of life among colorectal cancer survivors: results from the profiles 11 12 registry. Psychooncology. 2015;24(4):478-86. 13 14 15. HAHN E A, CELLA D, DOBREZ D G, et al. The impact of literacy on health-related quality of life 15 16 measurement and outcomes in cancer outpatients[J]. Qual Life Res, 2007,16(3):495-507. 17 18 16. Peterson PN, ShetterlyFor SM, peerClarke CL,et reviewal. Health Literacy onlyand Outcomes Among Patients With Heart 19 20 Failure. JAMA. 2011;305(16):1695–1701. 21 22 17. Wallace LS, Rogers ES, Roskos SE, et al. Brief report: screening items to identify patients with limited health 23 24 literacy skills. J Gen Intern Med. 2006;21(8):874-7. 25 26 18. Liu Z, Yang J. Health related quality of life: is it another comprehensive evaluation indicator of Chinese medicine 27 28 on acquired immune deficiency syndrome treatment? Journal of Traditional Chinese Medicine. 2015; 35(5):600-5. 29 30 19. Guyatt GH, Feeny DH, Patrick DL. Measuring Health-Related Quality-of-Life. Ann Intern Med. 1993; 31 32 118(8):622-9. 33 34 20. Wan CH, Chen QM, Zhang CZ, etc. Evaluation of the EORTC QLQ-C30 Chinese version of the cancer 35 36 patient's quality of life measurement scale [J]. Journal of Practical Oncology, 2005(04):353-355.
37 http://bmjopen.bmj.com/ 38 21. Meng Q, Wan CH, Luo JH, etc. Study on psychometrical characteristics of different scales among instruments 39 40 measuring quality of [J]. Tumor, 2011,31(03):245-249. 41 42 22. Song L, Mishel M, Bensen JT, et al. How does health literacy affect quality of life among men with newly 43 44 diagnosed clinically localized prostate cancer? Findings from the North Carolina-Louisiana Prostate Cancer
45 on September 23, 2021 by guest. Protected copyright. 46 Project (PCaP). Cancer. 2012;118(15):3842-51. 47 48 23. Chew LD, Griffin JM, Partin MR, et al. Validation of screening questions for limited health literacy in a 49 50 51 large VA outpatient population. J Gen Intern Med. 2008; 23(5), 561-566. 52 53 24. Wallace LS, Rogers ES, Roskos SE, et al. Brief report: screening items to identify patients with limited 54 55 health literacy skills. J Gen Intern Med. 2006;21(8):874-7. 56 57 25. Wallace LS, Cassada DC, Rogers ES, et al. Can screening items identify surgery patients at risk of limited 58 59 health literacy? J Surg Res. 2007;140(2):208-13. 60 13
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 23
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 26. Chew LD, Bradley KA, Boyko EJ. Brief questions to identify patients with inadequate health literacy. Fam 5 6 Med. 2004;36(8):588-94. 7 8 27. Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization for Research and Treatment of 9 10 Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl 11 12 Cancer Inst. 1993;85(5):365-76. 13 14 28. Wang JW. Tang Z, Yu JM, et al. QLQ-BR23 and EORTC+QLQ-C30 for measurement of the impact of 15 16 rehabilitation exercise on quality of life in breast cancer patients. Fudan Univ J Med Sci, 2015. 17 18 29. Fayers PM, AaronsonFor NK, Bjordalpeer K. The reviewEORTC QLQ-C30 scoringonly manual. 2001. 19 20 30. Osoba D, Rodrigues G, Myles J. Interpreting the significance of changes in health-related quality-of-life 21 22 scores. J Clin Oncol. 1998;16(1):139-44. 23 24 31. Davis TC, Long SW, Jackson RH, et al. Rapid estimate of adult literacy in medicine: a shortened screening 25 26 instrument. Fam Med. 1993;25(6):391-5. 27 28 32. Echeverri M, Anderson D, Nápoles AM. Cancer Health Literacy Test-30-Spanish (CHLT-30-DKspa), a New 29 30 Spanish-Language Version of the Cancer Health Literacy Test (CHLT-30) for Spanish-Speaking Latinos. J 31 32 Health Commun. 2016;21 Suppl 1:69-78. 33 34 33. Yehle KS, Plake KS, Nguyen P, et al. Health-Related Quality of Life in Heart Failure Patients With Varying 35 36 Levels of Health Literacy Receiving Telemedicine and Standardized Education. Home Healthc Now.
37 http://bmjopen.bmj.com/ 38 2016;34(5):267-72. 39 40 34. Baker DW, Williams MV, Parker RM, et al. Development of a brief test to measure functional health literacy. 41 42 Patient Educ Couns. 1999;38(1):33-42. 43 44 35. Halverson JL, Martinez-Donate AP, Palta M, et al. Health Literacy and Health-Related Quality of Life
45 on September 23, 2021 by guest. Protected copyright. 46 Among a Population-Based Sample of Cancer Patients. J Health Commun. 2015;20(11):1320-9. 47 48 36. Yao H, Shi Q, Li Y. The Current Status of Health Literacy in China. China Population Today. 2016; 41-41. 49 50 51 37. Huang L, Peng S, University FM, et al. The Status and Influencing Factors of Health Literacy among Cancer 52 53 Patients. Nursing Journal of Chinese Peoples Liberation Army. 2016. \ 54 55 38. http://www.shcrc.cn/webs/intro.aspx. Published 2012. Accessed February 1, 2018. 56 57 39. JANSEN L, KOCH L, BRENNER H, et al. Quality of life among long-term (>/=5 years) colorectal cancer 58 59 survivors--systematic review[J]. Eur J Cancer, 2010,46(16):2879-2888. 60 14
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 23 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 40. HAMASHIMA C. Long-term quality of life of postoperative rectal cancer patients[J]. J Gastroenterol Hepatol, 5 6 2002,17(5):571-576. 7 8 41. RAMSEY S D, KRISTIN B, CAROL M, et al. Quality of life in long term survivors of colorectal cancer.[J]. 9 10 American Journal of Gastroenterology, 2002,97(5):1228-1234. 11 12 42. SAPP A L, TRENTHAM-DIETZ A, NEWCOMB P A, et al. Social networks and quality of life among 13 14 female long-term colorectal cancer survivors[J]. Cancer, 2003,98(8):1749-1758. 15 16 43. TRENTHAM-DIETZ A, REMINGTON P L, MOINPOUR C M, et al. Health-related quality of life in female 17 18 long-term colorectalFor cancer survivors[J].peer Oncologist,review 2003,8(4):342-349. only 19 20 44. KIM S P, KNIGHT S J, TOMORI C, et al. Health literacy and shared decision making for prostate cancer patients 21 22 with low socioeconomic status[J]. Cancer Invest, 2001,19(7):684-691. 23 24 45. Protheroe J, Wolf MS, Lee A. Health literacy and health outcomes. Health Literacy in Context: International 25 26 Perspectives. 2010;57-74. 27 28 46. Aaby A, Friis K, Christensen B, et al. Health literacy is associated with health behaviour and self-reported 29 30 health: A large population-based study in individuals with cardiovascular disease. Eur J Prev Cardiol. 31 32 2017;24(17):1880-1888. 33 34 35 36 Figure Title:
37 http://bmjopen.bmj.com/ 38 Figure 1. Item Characteristic Curve and Test Information Curve for the 3 established screening questions 39 40 Figure 2. Unadjusted QOL scores for cancer survivors with inadequate and adequate health literacy 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 15
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 23
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Table 1. Characteristics of respondent stratified by HL levels* 5 6 Level of Health Literacy 7 8 Characteristics Total (N=4589) Adequate (n=4430) Inadequate (n=159) P value 9 10 Age, M (Q1,Q3), y 62.0 (57.04,66.53) 62.0 (57.00,66.53) 62.52 (58.37,68.39) 0.0485 11 12 Years with cancers, M (Q1,Q3) 6.09 (3.50,10.50) 6.09 (3.50,10.50) 5.92 (3.50,10.58) 0.6258 13 14 Sex, No.(%) 0.9424 15 16 Male 1057 (23.03) 1020 (96.50) 37 (3.50) 17 18 Female For peer3532 (76.97) review3410 (96.55) only122 (3.45) 19 20 Race 0.8644 21 22 Hanzu 4544 (99.28) 4386 (96.52) 158 (3.48) 23 24 Marriage 0.2192 25 26 single 129 (2.81) 121 (93.80) 8 (6.20) 27 28 married/ cohabitation 3989 (86.93) 3853 (96.61) 136 (3.41) 29 30 divorced/widowed/ separated 471 (10.226) 456 (96.82) 15 (3.18) 31 32 Annual income, ¥ 0.0534 33 34 <5000 4134 (91.89) 3987 (96.44) 147 (3.56) 35 36 ≥5000 365 (8.11) 359 (98.36) 6 (1.64)
37 http://bmjopen.bmj.com/ 38 Family income 0.8296 39 40 <5000 3086 (70.07) 2979 (96.53) 107 (3.47) 41 42 43 ≥5000 1318 (29.93) 1274 (96.66) 44 (3.34) 44 Education 0.0003
45 on September 23, 2021 by guest. Protected copyright. 46 47 Middle school or less 4021 (87.62) 3867 (96.17) 154 (3.83) 48 49 High school or college 568 (12.38) 563 (99.12) 5 (0.88) 50 51 Insurance status 0.8159 52 53 Uninsured 645 (15.55) 623 (96.59) 22 (3.41) 54 55 Medicare 3504 (84.45) 3378 (96.40) 126 (3.60) 56 57 Treatment regimen 0.2753 58 59 Chemotherapy 661 (14.44) 641 (96.97) 20 (3.03) 60 16
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 23 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Radiotherapy 74 (1.62) 68 (91.89) 6 (8.11) 5 6 Operation 931 (20.34) 902 (96.89) 29 (3.11) 7 8 Chemotherapy & Radiotherapy 162 (3.54) 155 (95.68) 7 (4.32) 9 10 Chemotherapy & Radiotherapy 897 (19.59) 870 (96.99) 27 (3.01) 11 12 & Operation 13 14 Chemotherapy & Operation 1658 (36.22) 1594 (96.14) 64 (3.86) 15 16 Biotherapy 69 (1.51) 66 (95.65) 3 (4.35) 17 18 Number of Chronic diseaseFor peer review only 0.3564 19 20 0 1238 (26.98) 1203 (97.17) 35 (2.83) 21 22 1-3 2735 (59.60) 2634 (96.31) 101 (3.69) 23 24 >3 616 (13.42) 593 (96.27) 23 (3.73) 25 26 Smoking habit 0.5284 27 28 Never 3841 (83.70) 3711 (96.62) 130 (3.38) 29 30 Former 598 (13.03) 573 (95.82) 25 (4.18) 31 32 Current 150 (3.27) 146 (97.33) 4 (2.67) 33 34 Current alcohol use 0.0637 35 36 None 4264 (92.92) 4113 (96.46) 151 (3.54)
37 http://bmjopen.bmj.com/ 38 Light to moderate 215 (4.69) 213 (99.07) 2 (0.93) 39 40 Heavy 110 (2.40) 104 (94.55) 6 (5.45) 41 42 Body mass index# 0.4593 43 44 <18.5 233 (5.16) 229 (98.28) 4 (1.72)
45 on September 23, 2021 by guest. Protected copyright. 46 18.5-23.9 2342 (51.85) 2258 (96.41) 84 (3.59) 47 48 24.0-27.9 1546 (34.23) 1492 (96.51) 54 (3.49) 49 50 51 ≥28.0 396 (8.77) 380 (95.96) 16 (4.04) 52 53 Physical activity 0.0264 54 55 No 1038 (22.64) 991 (95.47) 47 (4.53) 56 57 1-4 times per week 2493 (54.37) 2406 (96.51) 87 (3.49) 58 59 ≥5 times per week 1054 (22.99) 1029 (97.63) 25 (2.37) 60 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 23
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 * Data are given as percentages unless otherwise indicated. 5 6 # Calculated as weight in kilograms divided by the square of height in meters. 7 8 9 10 Table 2. Relationship between survivors’ characteristics and QOL 11 12 Unadjusted Adjusted* 13 Predictor 14 Coefficientτ P Value Coefficientτ P Value 15 16 HL score -2.11 <0.0001 -2.07 <0.0001 17 18 Age For -0.25peer review<0.0001 only-0.16 0.0053 19 20 Sex, female -3.92 <0.0001 -5.02 0.0002 21 22 Race/ethnicity 2.10 0.5854 6.41 0.1819 23 24 Marriage -3.01 0.0051 -2.14 0.0710 25 26 Education -0.28 0.6316 -1.38 0.3190 27 28 Annual income 0.42 0.7664 0.81 0.6044 29 30 Family income -0.03 0.9694 0.53 0.5521 31 32 Insurance status -1.67 0.1263 -0.09 0.9346 33 34 Years with cancer (per year) -0.23 0.0002 -0.04 0.5401 35 36 Number of Chronic disease -5.79 <0.0001 -4.87 <0.0001
37 http://bmjopen.bmj.com/ 38 Smoking habit 1.35 0.015 -1.23 0.1210 39 40 Alcohol intake 3.38 <0.0001 2.07 0.0240 41 42 43 Physical activity 4.07 <0.0001 3.15 <0.0001 44 Treatment regimen -0.004 0.985 0.16 0.4474
45 on September 23, 2021 by guest. Protected copyright. 46 47 District 0.63 <0.0001 0.48 <0.0001 48 49 BMI 1.14 0.0305 2.18 <0.0001 50 * Adjusted for age, sex, race, marriage, income, education, insurance, years with cancer, treatment regimen, district, 51 52 number of chronic disease, smoking habit, alcohol use, physical activity and BMI. 53 τAll coefficients correspond to a change in QOL score for unit change of each covariate. 54 55 56 57 58 59 60 18
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 23 BMJ Open
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 Table 3. Adjusted Odds Ratio of QOL for cancer survivors with inadequate vs adequate health literacy* 5 6 QOL Survivors with poor QOL Odds Ratio (95%CI) P value 7 8 Global health status (QL) 788 2.81 (1.94-4.06) <0.0001 9 10 Physical function (PF) 83 2.84 (1.18-6.87) 0.0205 11 12 Role function (RF) 53 2.70 (0.94-7.80) 0.0664 13 14 Emotional function (EF) 77 5.01 (2.37-10.58) <0.0001 15 16 Cognitive function (CF) 128 4.07 (2.18-7.61) <0.0001 17 18 Social function (SF) For peer250 review3.59 only (2.19-5.88) <0.0001 19 20 Fatigue (FA) 425 2.63 (1.68-4.12) <0.0001 21 22 Nausea/Vomiting (NV) 52 5.39 (2.17-13.38) 0.0003 23 24 Pain (PA) 296 2.37 (1.41-3.97) 0.0011 25 26 Dyspnea (DY) 152 1.86 (0.88-3.94) 0.106 27 28 Insomnia (SL) 475 2.41 (1.56-3.74) <0.0001 29 30 Appetite loss (AP) 91 3.84(1.84-8.03) 0.0003 31 32 Constipation (CO) 188 1.06 (0.45-2.46) 0.8995 33 34 Diarrhea (DI) 114 2.11 (0.95-4.71) 0.0679 35 36 Financial difficulties (FI) 747 2.73 (1.88-3.97) <0.0001
37 http://bmjopen.bmj.com/ 38 *Adjusted for age, sex, race, marriage, income, education, insurance, years with cancer, treatment regimen, district, number of chronic 39 disease, smoking habit, alcohol use, physical activity and BMI 40 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 19
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 23
1 2
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 Item 1 Item 2 6 Slope=1.49 Slope= 2.21 7 Threshold1= -1.03 Threshold1= -0.15 Threshold2= 0.81 Threshold2= 1.21 8 Threshold3= 2.48 Threshold3= 2.02 9 Threshold4= 3.80 Threshold4= 2.48 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 Item 3 24 Slope= 0.88 25 Threshold1= -1.81 Threshold2= -0.05 26 Threshold3= 2.47 27 Threshold4= 3.94 28 29 30 31 32 33 34 35 36
37 http://bmjopen.bmj.com/ 38 39 Figure 1. Item Characteristic Curve and Test Information Curve for the 3 established screening questions 40 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 23 BMJ Open
1 2 3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 2. Unadjusted QOL scores for cancer survivors with inadequate and adequate health literacy 33 34 35 36
37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 23
1 2 STROBE Statement—Checklist of items that should be included in reports of cross-sectional studies
3 BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 Item 6 No Recommendation 7 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or 8 the abstract (Page 2, Line 1) 9 10 (b) Provide in the abstract an informative and balanced summary of what 11 was done and what was found (Page 2, Line 17) 12 Introduction 13 14 Background/rationale 2 Explain the scientific background and rationale for the investigation being 15 (Page 3, Line 9-Page 4, Line 16 ) reported 16 Objectives (Page 4, Line 17-22) 3 State specific objectives, including any prespecified hypotheses 17 18 Methods For peer review only 19 Study design (Page 4, Line 26) 4 Present key elements of study design early in the paper 20 Setting (Page 4, Line 26-Page 5, 5 Describe the setting, locations, and relevant dates, including periods of 21 22 Line 1) recruitment, exposure, follow-up, and data collection 23 Participants (Page 5, Line 2-13) 6 (a) Give the eligibility criteria, and the sources and methods of selection of 24 participants 25 Variables (Page 4, Line 18-Page 7 Clearly define all outcomes, exposures, predictors, potential confounders, 26 27 6, Line 10) and effect modifiers. Give diagnostic criteria, if applicable 28 Data sources/ measurement 8* For each variable of interest, give sources of data and details of methods 29 (Page 4, Line 18-Page 5, Line 7) of assessment (measurement). Describe comparability of assessment 30 methods if there is more than one group 31 32 Bias (Page 10, Line 1) 9 Describe any efforts to address potential sources of bias 33 Study size 10 Explain how the study size was arrived at 34 (Page 5, Line 10-13) 35 36 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If (Page 5, Line 18-20) applicable, describe which groupings were chosen and why
37 http://bmjopen.bmj.com/ 38 Statistical methods 12 (a) Describe all statistical methods, including those used to control for 39 (Page 6, Line 11-25) confounding 40 41 (b) Describe any methods used to examine subgroups and interactions 42 (c) Explain how missing data were addressed 43 (d) If applicable, describe analytical methods taking account of sampling 44 strategy
45 on September 23, 2021 by guest. Protected copyright. 46 (e) Describe any sensitivity analyses 47 Results 48 49 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers 50 (Page 7, Line 1-3) potentially eligible, examined for eligibility, confirmed eligible, included 51 in the study, completing follow-up, and analysed 52 (b) Give reasons for non-participation at each stage 53 54 (c) Consider use of a flow diagram 55 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, 56 (Page 7, Line 4-14) social) and information on exposures and potential confounders 57 (b) Indicate number of participants with missing data for each variable of 58 59 interest 60 Outcome data 15* Report numbers of outcome events or summary measures Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted
1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 23 BMJ Open
1 2 (Page 7, Line 15- Page 8, Line 6) estimates and their precision (eg, 95% confidence interval). Make clear 3 which confounders were adjusted for and why they were included BMJ Open: first published as 10.1136/bmjopen-2018-028458 on 31 December 2019. Downloaded from 4 5 (b) Report category boundaries when continuous variables were 6 categorized 7 (c) If relevant, consider translating estimates of relative risk into absolute 8 risk for a meaningful time period 9 10 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, 11 and sensitivity analyses 12 Discussion 13 14 Key results 18 Summarise key results with reference to study objectives 15 (Page 8, Line 8-12) 16 Limitations 19 Discuss limitations of the study, taking into account sources of potential 17 18 (Page 10, Line 1-8) For peerbias or imprecision.review Discuss bothonly direction and magnitude of any potential 19 bias 20 Interpretation 20 Give a cautious overall interpretation of results considering objectives, 21 (Page 8, Line 13-Page 9, Line limitations, multiplicity of analyses, results from similar studies, and other 22 23 19) relevant evidence 24 Generalisability 21 Discuss the generalisability (external validity) of the study results 25 (Page 9, Line 20-26) 26 27 Other information 28 Funding 22 Give the source of funding and the role of the funders for the present study 29 (Page 10, Line 23) and, if applicable, for the original study on which the present article is 30 based 31 32 33 *Give information separately for exposed and unexposed groups. 34 35 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 36 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely
37 http://bmjopen.bmj.com/ 38 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 39 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 40 available at www.strobe-statement.org. 41 42 43 44
45 on September 23, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml