Prevalence and factors associated with type 2 diabetes mellitus and hypertension among the hill tribe elderly populations in northern Thailand

Prevalence and factors associated with type 2 diabetes mellitus and hypertension among the hill... Background: Type 2 diabetes mellitus (T2DM) and hypertension (HT) are major noncommunicable health problems in both developing and developed countries, including Thailand. Each year, a large amount of money is budgeted for treatment and care. Hill tribe people are a marginalized population in Thailand, and members of its elderly population are vulnerable to health problems due to language barriers, lifestyles, and daily dietary intake. Methods: An analytic cross-sectional study was conducted to estimate the prevalence of T2DM and HT and to assess the factors associated with T2DM and HT. The study populations were hill tribe elderly adults aged ≥ 60 years living in Chiang Rai Province, Thailand. A simple random method was used to select the targeted hill tribe villages and participants into the study. A validated questionnaire, physical examination form, and 5-mL blood specimen were used as research instruments. Fasting plasma glucose and blood pressure were examined and used as outcome measurements. Chi-square tests and logistic regression were used for detecting the associations between variables at the significance level alpha=0.05. Results: In total, 793 participants participated in the study; 49.6% were male, and 51.7% were aged 60-69 years. A total of 71.5% were Buddhist, 93.8% were uneducated, 62.9% were unemployed, and 89 % earned an income of < 5,000 baht/month. The overall prevalence of T2DM and HT was 16.8% and 45.5%, respectively. Approximately 9.0% individuals had comorbidity of T2DM and HT. Members of the Lahu, Yao, Karen, and Lisu tribes had a greater odds of developing T2DM than did those of the Akha tribe. Being overweight, having a parental history of T2DM, and having high cholesterol were associated with T2DM development. In contrast, those who engaged in highly physical activities and exercise had lower odds of developing T2DM than did those who did not. Regarding HT, being female, having a high dietary salt intake, being overweight, and having a parental history of HT were associated with HT development among the hill tribe elderly populations. Conclusions: The prevalence of T2DH and HT among the hill tribe elderly populations is higher than that among the general Thai population. Public health interventions should focus on encouraging physical activity and reducing personal weight, dietary salt intake, and greasy food consumption among the hill tribe elderly. Keywords: Type 2 diabetes mellitus, Hypertension, Hill tribe, Elderly, Thailand Correspondence: tk2016ms@gmail.com; tawatchai.api@mfu.ac.th Center of Excellence for the Hill tribe Health Research, Mae Fah Luang University, Chiang Rai, Thailand School of Health Science, Mae Fah Luang University, Chiang Rai, Thailand © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Apidechkul BMC Public Health (2018) 18:694 Page 2 of 17 Background the number of deaths was 21,830 cases (9,430 males, Type 2 diabetes mellitus (T2DM) and hypertension 12,400 females) [10]. The average cost of each T2DM case (HT) are common noncommunicable diseases among in attending hospital services per year was 598US$ for an elderly adults aged ≥ 60 years in both developing and de- independent case and 2,700US$ for a disabled case. veloped countries [1]. The prevalence of T2DM and HT Therefore, Thailand spends a large amount of money on varies according to age, sex, and race [2, 3]. There are the health care system annually [11]. different factors associated with T2DM and HT in differ- High blood pressure is a key risk factor for many ent populations, particularly among those with different diseases, including heart attack and stroke. In 2017, lifestyles and cultures [3, 4]. Older populations are the WHO estimated that more than one billion people most vulnerable to the development of T2DM and HT had HT caused 12.8% of all deaths and accounted for [5, 6]. T2DM and HT have become major causes of 57 million disability-adjusted life years (DALYs) or a morbidity and mortality of elderly populations in all total of 3.7% DALYs every year [12]. Thailand re- countries [7, 8]. The impact of T2DM and HT is not ported that 29.0% of adult Thais had HT, and only limited to physical and mental consequences; rather, it 37.0% for those people who had been diagnosed had also affects family and national economics [9]. Health their blood pressure under control in 2017 [13]. The professionals in health care institutes must manage the number of resistant HT patients in all health insti- maintenance of plasma glucose levels among T2DM pa- tutes in the entire country has increased from tients and blood pressure among HT patients using dif- 3,946,902 cases in 2013 to 5,584,007 cases in 2017 ferent regiments of drugs for their entire lives. With [13]. The statistics represent the full picture of the these demands, there are required numbers of health situation in Thailand, but there is no information professionals and large amount of financial input needed available on any specific subgroup of populations, to operate the treatment and care system each year. such as the hill tribe population. Patients need to frequently attend a clinic to meet and The hill tribe people are those who have migrated receive care from a doctor. Otherwise, many complica- from the southern region of China to Thailand in the tions could possibly develop, resulting in intensive and past century [14]. They are divided into six different complicated methods of treatment and care. main groups: Akha, Lahu, Karen, Hmong, Yao, and In 2014, the WHO estimated that 422 million people Lisu [15]. Approximately 2.5 million of the hill tribe worldwide were suffering from T2DM, which accounted people were living in Thailand in 2017 [16]. They for 8.5% of the prevalence among people over 18 years have their own culture, language and lifestyles, par- old. The prevalence is increasing among people aged > 30 ticularly in daily cooking. Some tribes use a high vol- years old, particularly in low- and middle-income coun- ume of oil for cooking, whereas other tribes use a tries. People aged ≥ 60 years old are also commonly de- high volume of salt for their daily food [14, 17]. Most fined as a vulnerable population for T2DM [2]. of them have similar cultural patterns in terms of Commonly, T2DM is a disease that progresses slowly using alcohol, particularly for religious rituals [18]. from its onset, and it may be diagnosed several years later. In 2017, the hill tribe elderly populations lived ac- T2DM is a major cause of other health problems, such as cording to their own traditional lifestyle and living blindness, kidney failure, heart attacks, stroke, and lower environment. They consumed drinks and foods pre- limb amputation. The WHO also reported that 1.6 million pared traditionally. Individual health care was mainly deaths were directly caused by diabetes, and almost half of based on their local healing patterns. With the prob- all deaths attributable to high blood glucose occurred be- lems of distance, language and discrimination, their fore the age of 70 years [2]. This finding reflects the need access to the Thai health care system was poor [19]. to regularly investigate those vulnerable to an early diag- Therefore, access to modern medical care is not nosis and determine ways of obtaining a better prognosis. common, especially for those who live very far from In 2016, the total T2DM prevalence among the Thai the city. Ultimately, the findings of the study could population was 9.6%: 9.1% in males and 10.1% in females. support the development of the health care service The total number of deaths caused by T2DM was 20,570 system for the hill tribe elderly populations. The cases; in the 30-69 year age group, the number of findings could also be used for the development of deaths was 8,120 cases (3,610 males, 4,510 females) and DM and HT prevention and control measures in in the ≥ 70 years age group, the number of deaths was these populations. Currently, there is no available in- 12,450 cases (4,760 males, 7,690 females). Moreover, formation about T2DM and HT among these popula- total number of deaths attributable to high blood tion groups. Therefore, the study aimed to estimate glucose was 35,640 cases; in the 30-69 year age the prevalence and factors associated with DM and group, the number of deaths was 13,810 cases (7,220 HT among the hill tribe elderly populations in north- males, 6,590 females) and in the ≥ 70 years age group, ern Thailand. Apidechkul BMC Public Health (2018) 18:694 Page 3 of 17 Methods Sample selection and preparing the participants Study design and participants The list of the hill tribe villages in Chiang Rai Province Study design was requested from the Hill Tribe Welfare and Develop- A cross-sectional study was conducted to gather infor- ment Center in Chiang Rai [21]. There were 749 hill mation from the selected subjects. tribe villages in Chiang Rai, which breakdown into 316 Lahu villages, 243 Akha villages, 63 Yao villages, 56 Hmong villages, 36 Karen villages, and 35 Lisu villages. Study setting In 2016, a total of 41,366 hill tribe families lived in The study was conducted along 16 districts in Chiang Chiang Rai Province. Rai Province, which is located in Thailand. Permission to access the villages had been granted by the District Government Officer. Sixty hill tribe villages, Study population or 10 villages in each tribe, were selected by a simple The study population was comprised of hill tribe elderly random method. A village headman was contacted and adults aged ≥ 60 years old who had lived in the study informed of all essential information regarding the re- setting for at least 3 years. search objective and its protocol. The list of elderly people who met the inclusion and exclusion criteria in the village was sent to the researcher. A simple random Eligible population method was used to select 13 individuals in each village, Elderly adults with the following characteristics were eli- after which they were invited to participate in the study. gible for the study: a) being classified as a member of After informing the village headman about the research the hill tribe by verbal confirmation, b) being ≥ 60 years objectives and protocols, some tribes collected more old, c) living in the study area for 3 years at the date of than the minimum required sample size: Lahu (an excess data collection, and d) having the ability to provide es- of 3 participants) and Hmong (an excess of 10 partici- sential information. Those who had been diagnosed with pants). Those who agreed to join the project were in- type 1 diabetes mellitus, which requires daily administra- formed of all research processes, including the tion of insulin, were excluded from the study. preparation of NPO (nothing per oral) for at least 8 hours for the blood specimen collection on the next day (Fig. 1). Sample size Six research assistants fluent in Thai and in one of The sample size was calculated by Epi-Info version 7.2 the six hill tribal languages were recruited. Selected re- (US Centers for Disease Control and Prevention, At- search assistants were trained in procedures, and the lanta, GA). By setting the alpha error at 0.05, the power required documents were completed three days before at 0.8, the previous prevalence of T2DM among the working in the field. Most of the hill tribe elderly exposed group at 18.0%, and the prevalence among the populations do not speak Thai. Therefore, there was a unexposed group at 0.07% [20], the sample size was cal- need to obtain complete information using the culated to be a minimum of 705 participants. Increasing research assistants. Recruiting young adults to help as the sample size by 10.0% for error resulted in 775 partic- research assistants was possible because hill tribe ipants required. community members younger than 25 years old had Since the sample size was calculated at 775 participants, already completed secondary level education in Thai at least 130 participants were needed in each tribe. schools. 749 hill tribe villages 10 villages/tribe 13-14 persons/village 130 Akhas 133 Lahus 140 Hmongs 130 Yaos 130 Karens 130 Lisus Fig. 1 Flowchart of participants’ selection. Sixty hill tribe villages were randomly selected from 749 villages, and 13-14 elderly people in each village were recruited into the study Apidechkul BMC Public Health (2018) 18:694 Page 4 of 17 Measurement Blood pressure was assessed twice in all participants Research instruments with a 15-minute gap between assessments, and in case A questionnaire, physical examination form, a manual systolic and/or diastolic blood pressures were greater sphygmomanometer, and a 5-mL blood specimen served than 90 mmHg and/or 150 mmHg, respectively, it was nd as research instruments. A questionnaire was developed assessed again after 15 minutes of rest following the 2 from the review literature. After completion of the first assessment. Participants with 90 mmHg and/or 140 draft, the validity was detected by the item-objective mmHg of systolic and diastolic blood pressures, respect- congruence (IOC) technique in which three external ex- ively, were diagnosed as HT patients [22]. perts in relevant fields verified the validity. Questions Body mass index (BMI) was classified into three cat- with scores less than 0.5 were excluded, those with egories, according to the WHO guidelines for Asian scores 0.5-0.69 were revised, and those with scores populations: underweight (BMI≤18.5), normal weight greater than 0.7 were defined as acceptable to use. The (BMI=18.51-22.99), and overweight (BMI ≥ 23.00) [23]. questionnaire was also tested for reliability by pilot test- A 5-mL blood specimen was collected from a peripheral ing it in 15 similar participants in the Ban San Ti Suk vein puncture. After blood was drawn, a 3-mL blood spe- hill tribe village using the test-retest method. Questions cimen was collected and stored in a sodium fluoride tube with Cronbach’s alpha ≥ 0.5 were included in the form. to detect fasting plasma glucose. Another 2-mL blood spe- Ultimately, there were 28 questions in three parts in- cimen was collected and stored in a clot blood clot tube cluded in the questionnaire, which presented an overall for detecting lipid profiles. Uric acid, cholesterol, and tri- Cronbach’s alpha of 0.77. glycerides were assessed in mg/dL. Participants with uric In the first parts, 13 questions were used to collect acid ≥ 7 mg/dL, cholesterol ≥ 200 mg/dL, and triglycerides participants’ general information, such as age, sex, edu- ≥ 200 mg/dL were defined as a high-level group [24]. Par- cation, and religion. Fifteen questions were included in ticipants with fasting plasma glucose ≥ 126 mg/dL were the second part, including questions about health behav- asked to provide another blood specimen within a week to iors such as “Do you smoke?”, “Do you drink alcohol?”, determine type 2 diabetes stage. and “Do you use methamphetamine?”. In these ques- tions, the three answer choices were “Yes”, “Ever in the Procedures past”, and “No”. Data gathering procedures Several questions were asked regarding daily food con- After the consent form was obtained, a 5-mL blood spe- sumption and exercise, such as “Do you usually eat a salty cimen was collected. Participants were asked to diet?”, “Do you favor having greasy food?”,and “Do you complete the questionnaire in a private room in the vil- like to eat sweet food?” Two answer choices were provided lage with the help of the research assistants. A trained for the questions, “Yes” and “No”. However, for exercise physician examined the physical health of all participants practices, the three answer choices were “No”, “Highly ac- in a proper room. A small gift was given to participants tive physical work, such as farmer and labor”,and “Yes”. after they completed the questionnaire. Questions on medications, history of T2DM and HT, and parental history of T2DM and HT were also included. Statistical analysis To confirm the diagnosis, all participants who responded Descriptive statistics, such as the means, minimums, that they had T2DM and HT were asked to present the maximums, standard deviations, and percentages, were log-book from a hospital. In Thailand, all DM and HT used to explain the general characteristics of the partici- cases are provided individual log-books to use to collect pants. Chi-square tests and logistic regressions were medical information and make appointments. used to detect the associations between variables at the The last part consisted of twenty-one items of a phys- significance level α=0.05. Logistic regression was used to ical examination form, which is used at Mae Fah Luang detect the associations between variables in both univar- University Hospital. This form resembles a checklist and iate and multivariate models. The “ENTER” mode was can include more information if required. A manual used to select the significant variables in the model. The sphygmomanometer was used for assessing blood significance level (alpha) was set at 0.05 in both univari- pressure. ate and multivariate analyses. Variables that were found to be significant in the univariate analysis were retained Variables and measurements in the multivariate analysis. In the multivariate model, T2DM in the study was identified by the following: a) the most nonsignificant variable was deleted from the having no history of a medical diagnosis, such as type 1 model before running the second step. The model was diabetes mellitus, at a particularly early age after birth, analyzed until all remaining variables were found to be b) having shown a fasting plasma glucose ≥ 126 mg/dL significant at an alpha level of 0.05, and the results were twice on different days [20]. interpreted. Apidechkul BMC Public Health (2018) 18:694 Page 5 of 17 Table 1 General characteristics of the study participants Table 1 General characteristics of the study participants (Continued) Characteristics Number Percent Characteristics Number Percent Total 793 100.0 Merchant 11 1.4 Sex Labor 19 2.4 Male 393 49.6 Other 12 1.5 Female 400 50.4 Monthly family income (baht) Thai ID card 0 69 8.7 Yes 745 93.9 ≤5,000 707 89.2 No 48 6.1 ≥5,001 17 2.1 Tribe Debt (baht) Akha 130 16.4 0 673 84.9 Lahu 133 16.8 ≤5,000 14 1.8 Hmong 140 17.6 5,001-10,000 11 1.4 Yao 130 16.4 10,001-50,000 58 7.3 Karen 130 16.4 ≥50,001 37 4.6 Lisu 130 16.4 Age (years) 60-69 410 51.7 Results 70-79 279 35.2 Characteristics of participants ≥ 80 104 13.1 In total, 793 participants were recruited into the study. Proportions of participants were mostly equal by sex and Religion among the six tribes. A few people had no Thai identifica- Buddhism 567 71.5 tion card (6.1%), with an equal proportion among the Christianity 225 28.4 tribes. The majority were aged 60-69 years (51.7%), with Islam 1 0.1 an average age of 70.1 years (range=60-100, SD=7.57, Education max=100, and min=60). The majority of the sample prac- None 739 93.8 ticed Buddhism (71.5%) and had no education (94.8%). A few people lived alone (6.1%), and most participants were Primary School 41 5.2 married (66.8%). Regarding economic status, 89.2% had an High School 8 1.0 income of ≤ 5,000 baht/month (mean=1,129 baht, Resides with SD=1,273), and 84.9% had no debt (Table 1). Child 559 70.5 There were no statistical differences in the distribution Cousin 12 1.5 of participants according to sex and tribe in three differ- Spouse 174 21.9 ent age categories (60-69, 70-79, and ≥ 80 years). A few of the hill tribe elderly adults had the ability to commu- Alone 48 6.1 nicate in Thai: 19.5% could speak, 19.5% could under- Marital status stand, 2.0% could read, and 1.6% could write fluently. Single 15 1.9 Males had significantly better Thai communication skills Married 524 66.8 than females in all four domains: speaking, understand- Divorced 20 2.5 ing, reading, and writing. Widow 226 28.8 The prevalence of T2DM and HT was 16.8% and 45.5%, respectively. Seventy-five participants had been Number of family member (persons) diagnosed with T2DM before being recruited into the 1 40 5.0 study. Among these participants, 8 (10.6%) had high 2 116 14.6 fasting glucose or were unable to control blood glucose 3-5 301 38.0 after medication. Fifty-five participants (7.7%) were de- 6 336 42.4 tected as new T2DM cases (Table 2). However, 18 par- Occupation ticipants (1.2%) could not draw blood specimens. Two hundred and forty participants (30.3%) had been Unemployed (retired) 499 62.9 diagnosed with HT, among whom 37.9% were unable to Farmer 252 31.8 control their blood pressure after medication. After Apidechkul BMC Public Health (2018) 18:694 Page 6 of 17 Table 2 Prevalence of T2DM and HT among the participants Table 3 Comparison of T2DM and HT by participants’ characteristics Chracteristics Number Percent 2 2 Characteristic T2DM χ p-value HT χ p-value Medical history of T2DM Yes No Yes No No 718 90.5 (%) (%) (%) (%) Yes 75 9.5 Sex Effective control of blood glucose by daily medication Male 66 316 0.13 0.712 164 229 4.52 0.034* (17.3) (82.7) (41.7) (58.3) No 8 10.6 Female 64 329 197 203 Yes 67 89.4 (16.3) (83.7) (49.3) (50.7) Fasting plasma glucose level among non-DM diagnosed Age (years) Normal 645 89.8 60-69 75 324 2.49 0.287 173 237 4.25 0.119 High (T2DM) 55 7.7 (18.8) (81.2) (42.2) (57.8) 70-79 39 234 134 145 (Missing=18, 2.5%) (14.3) (85.7) (48.0) (52.0) Prevalence of T2DM=16.8% ≥80 16 87 54 50 Medical history of HT (15.5) (84.5) (51.9) (48.1) No 553 69.7 Tribe Yes 240 30.3 Akha 11 117 24.48 <0.001* 61 69 26.45 <0.001* (8.6) (91.4) (46.9) (53.1) Effective control of blood pressure by daily medication Lahu 26 107 61 72 No 91 37.9 (19.5) (80.5) (45.9) (54.1) Yes 149 62.1 Hmong 11 124 42 98 (8.1) (91.9) (30.0) (70.0) Blood pressure level among non-HT diagnosed Yao 26 95 74 56 Normal 432 78.1 (21.5) (78.5) (56.9) (43.1) High (HT) 121 21.9 Karen 34 95 52 78 Prevalence of HT=45.5% (26.4) (73.6) (40.0) (60.0) Having both T2DM and HT 70 9.0 Lisu 22 107 71 59 (17.1) (82.9) (54.6) (45.4) The overall prevalence of T2DM among the participants The overall prevalence of HT among the participants *Significance level at α=0.05 those who had no history of HT diagnosis and medication With regard to the physical health and medical his- were seen, 121 participants (21.9%) were detected as new tory among the participants, 45.0% were overweight, HT cases. Finally, 70 cases (9.0%) were determined to have 6.8% were disabled persons, 15.0% had sleeping prob- both T2DM and HT: 36 males and 34 females (Table 2). lems, 9.7% had cataracts, 28.7% had hearing problems, There was statistical significance in the proportion of and 43.3% had tooth problems (Table 7). participants with T2DM and HT by sex and tribe. Only There were statistically significant differences in the qual- the participants with T2DM showed a statistically signifi- ity of uric acid and cholesterol according to sex, age cat- cant difference in proportion (Table 3). egory, and tribe. A greater proportion of males, individuals Health behaviors among the participants indicated that in higher age categories, and Lahu and Lisu tribe members 19.7% smoked, 14.6% drank alcohol, 44.9% ate uncooked hadhighuricacidlevelsthandid females, thoseinyounger food, 23.8% chewed tobacco, and 10.1% did not exercise age categories, and members of other tribes. Only age cat- regularly. A comparison of health behaviors such as smok- egory and tribe showed significant differences on the level ing, alcohol use, eating uncooked food, and regular exercise of triglycerides; a greater proportion of those in lower age among the tribes showed statistically significant differences categories had high cholesterol than those in higher age (Table 4). Additionally, there were significant sex differ- categories. A greater proportion of members of the Lahu ences in the following health behaviors: smoking; alcohol and Akha tribes were in the high cholesterol group com- use; the consumption of uncooked food, salty food, greasy pared to those in the remaining tribes (Table 8). food, and sweet food; opium use; chewing tobacco; and In themultivariatemodel, fivefactors were associated regular exercise (Table 5). with T2DM: tribe, exercise, BMI, parental history of Most participants had moderate levels of T2DM, and triglycerides. The Lahu, Yao, Karen, and health-related knowledge, attitudes, and practices. Lisu tribes had greater odds of developing T2DM than Only the distribution of attitudes by tribe showed the Akha tribe, with OR =2.89 (95%CI=1.32-6.33), adj statistical significance (Table 6). OR =3.47 (95%CI=1.58-7.62), OR =5.03 (95%CI= adj adj Apidechkul BMC Public Health (2018) 18:694 Page 7 of 17 Table 4 Characteristics of health behaviors by tribe Health behaviors Tribe χ p-value Total Akha Lahu Hmong Yao Karen Lisu n% n% n % n % n % n% n% Smoking No 486 61.3 94 19.3 70 14.4 106 21.8 72 14.8 48 9.9 96 19.8 79.02 < 0.001* Ever in the past 151 19.0 12 7.9 33 21.9 11 7.3 29 19.2 50 33.1 16 10.6 Yes 156 19.7 24 15.4 30 19.2 23 14.7 29 18.6 32 20.5 18 11.5 Alcohol use No 538 67.8 99 18.4 92 17.1 109 20.3 88 16.4 77 14.3 73 13.6 43.93 < 0.001* Ever 139 17.5 13 9.4 29 20.9 14 10.1 17 12.2 25 18.0 41 29.5 Yes 116 14.6 18 15.5 12 10.3 17 14.7 25 21.6 28 24.1 16 13.8 Methamphetamine use No 776 97.9 124 16.0 132 17.0 137 17.7 126 16.4 128 16.5 129 16.6 12.15 0.275 Ever in the past 2 0.3 0 0.0 0 0.0 0 0.0 1 50.0 1 50.0 0.0 0.0 Yes 15 1.9 6 40.0 1 6.7 3 20.0 3 20.0 1 6.7 1 6.7 Opium use No 723 91.2 112 15.5 125 17.3 124 17.2 115 15.9 123 17.0 124 17.2 15.77 0.106 Ever in the past 54 6.8 12 22.2 6 11.1 12 22.2 12 22.2 7 13.0 5 9.3 Yes 16 2.0 6 37.5 2 12.5 4 25.0 3 18.8 0 0.0 1 6.3 Eating uncooked food No 385 48.5 79 20.5 74 19.2 68 17.7 69 17.9 43 11.2 52 13.5 29.65 < 0.001* Ever in the past 52 6.6 5 9.6 6 11.5 9 17.3 8 15.4 11 21.2 13 25.0 Yes 356 44.9 46 12.9 53 14.9 63 17.7 53 14.9 76 21.3 65 18.3 Chewing No 604 76.2 70 11.6 108 17.9 135 22.4 128 21.2 100 16.6 63 10.4 159.80 < 0.001* Yes 189 23.8 60 31.7 25 13.2 5 2.6 2 1.1 30 15.9 67 35.4 Regular exercise No 80 10.1 22 27.5 7 8.8 15 18.8 7 8.8 22 27.5 7 8.8 37.50 < 0.001* Yes 433 54.6 68 15.7 88 20.3 75 17.3 66 15.2 54 12.5 82 18.9 Highly active physical work 280 35.3 40 14.3 38 13.6 50 17.9 57 20.4 54 19.3 41 14.6 *Significance level at α=0.05 2.35-10.78), and OR =2.73 (95%CI=1.22-6.07) respectively. dietary salt intake, BMI, and parental history of HT. adj Those who were overweight had greater odds of developing Females had greater odds of developing HT than males, T2DM than those with normal weight, with OR =2.08 with OR =1.29 (95%CI=1.01-1.68). Those who had adj adj (95%CI=1.32-3.27). Those who had a parental history of dietary salt intake had greater odds of developing HT T2DM had greater odds of developing T2DM than those than those who did not, with OR =1.48 adj who did not, with OR =1.55 (95%CI=1.17-2.10). Those (95%CI=1.14-2.00). Those who were overweight had adj with high cholesterol had greater odds of developing greater odds of developing HT than those with normal T2DM than those with low cholesterol, with OR =1.73 weight, with OR =1.37 (95%CI=1.01-1.90), and those adj adj (95%CI=1.10-2.73). Those who engaged in high levels of who had a parental history of HT had greater odds of physical activity and exercise had lower odds of developing developing HT than those who did not, with OR =3.38 adj T2DM than those who did not, with OR =0.48 (95%CI=2.81-4.48) (Table 10). adj (95%CI=0.25-0.91) and OR =0.45 (95%CI=0.24-0.83), re- adj spectively (Table 9). Discussion Four factors were found to be associated with HT after Members of the hill tribe elderly population are living controlling for all possible confounding variables: sex, with a high burden of T2DM and HT in Thailand. There Apidechkul BMC Public Health (2018) 18:694 Page 8 of 17 Table 5 Comparison of health behavior by sex Health behvaior Total Male Female χ p-value n % n% n% Smoking No 486 61.3 151 31.1 335 68.9 173.52 < 0.001* Ever in the past 151 19.0 125 82.8 26 17.2 Yes 156 19.7 117 75.0 39 25.0 Alcohol use No 538 67.8 169 31.4 369 68.6 222.02 < 0.001* Ever in the past 139 17.5 117 84.2 22 15.8 Yes 116 14.6 107 92.2 9 7.8 Consumption of uncooked food No 385 48.5 106 27.5 279 72.5 145.24 < 0.001* Ever in the past 52 6.6 37 71.2 15 28.8 Yes 356 44.9 250 70.2 106 29.8 Salty food No 282 35.6 106 37.6 176 62.4 25.05 < 0.001* Yes 511 64.4 287 56.2 224 43.8 Greasy food No 297 37.5 194 65.3 103 34.7 47.12 < 0.001* Yes 496 62.5 199 40.1 297 59.9 Sweet food No 391 49.3 216 55.2 175 44.8 9.96 0.0016* Yes 402 50.7 177 44.0 225 56.0 Opium use No 723 91.2 339 46.9 384 53.1 23.95 < 0.001* Ever in the past 54 6.8 43 79.6 11 20.4 Yes 16 2.0 11 68.8 5 31.3 Methamphetamine use No 776 97.9 381 49.1 395 50.9 3.69 0.079 Yes 17 2.1 12 70.6 5 29.4 Chewing No 604 76.2 313 51.8 291 48.2 5.19 0.023* Yes 189 23.8 80 42.3 109 57.7 Regular exercise No 433 54.6 184 42.5 249 57.5 26.05 < 0.001* Highly active physical work 280 35.3 173 61.8 107 38.2 Yes 80 10.1 36 45.0 44 55.0 *Significance level at α=0.05 are several factors associated with HT and T2DM, such read and write in Thai. The prevalence of T2DM and as behaviors related to daily living, culture and food HT was 16.8% and 45.5%, respectively, of which 7.7% practices. Most members of the hill tribe elderly popula- and 21.9% represented the incident rates for T2DM and tion have no education and low economic status. Very HT, respectively. Moreover, 9.3% of T2DM participants few have Thai ID cards, which is usually used to access and 37.9% of HT participants could not control their all public services in Thailand, including health care ser- plasma glucose and blood pressure after having daily vices [17]. Only one-fourth of the participants were able medication. The comorbidity rate was approximately to speak and understand Thai, and a few people could one-fourth of the participants who used alcohol and Apidechkul BMC Public Health (2018) 18:694 Page 9 of 17 Table 6 Comparison on knowledge, attitudes, and practices regarding health among tribes KAP Tribe χ p-value Total Akha Lahu Hmong Yao Karen Lisu n % n % n% n % n% n% n % Total 377 100.0 60 15.9 76 20.2 46 12.2 70 18.6 73 19.4 52 13.8 Knowledge Low 61 16.2 15 24.6 13 21.3 10 16.4 8 13.1 5 8.2 10 16.4 15.07 0.129 Moderate 167 44.3 24 14.4 33 19.8 21 12.6 38 22.8 31 18.6 20 12.0 High 149 39.5 21 14.1 30 20.1 15 10.1 24 16.1 37 24.8 22 14.8 Attitude Low 53 14.1 12 22.6 5 9.4 14 26.4 14 26.4 4 7.5 5 9.4 38.04 < 0.001* Moderate 250 66.3 44 17.6 55 22.0 25 10.0 42 16.8 44 17.6 40 16.0 High 74 19.6 4 5.4 16 21.6 7 9.5 14 18.9 25 33.8 8 10.8 Practice Low 47 12.5 3 6.4 8 17.0 10 21.3 9 19.1 8 17.0 9 19.1 10.51 0.397 Moderate 267 70.8 44 16.5 56 21.0 27 10.1 49 18.4 54 20.2 37 13.9 High 63 16.7 13 20.6 12 19.0 9 14.3 12 19.0 11 17.5 6 9.5 *Significance level at α=0.05 Table 7 Physical examination and medical history Item Total Male Female χ p-value n% n% n% BMI Underweight 116 14.6 62 53.4 54 46.6 3.98 0.137 Normal 320 40.4 168 52.5 152 47.5 Overweight 357 45.0 163 45.7 194 54.3 Disabled No 739 93.2 362 49.0 377 51.0 1.42 0.232 Yes 54 6.8 31 57.4 23 42.6 Heart disease No 724 96.1 337 46.5 387 53.5 0.37 0.538 Yes 29 3.9 16 55.2 13 44.8 History of TB diagnosis No 757 95.5 369 48.7 388 51.3 4.41 0.036* Yes 36 4.5 24 66.7 12 33.3 Sleeping problem No 674 85.0 356 52.8 318 47.2 19.09 < 0.001* Yes 119 15.0 37 31.1 82 68.9 Eye Normal 663 83.6 328 49.5 335 50.5 0.99 0.804 Cataract 77 9.7 36 46.8 41 53.2 Pterygium 50 6.3 27 54.0 23 46.0 History of glaucoma 3 0.4 2 66.7 1 33.3 Apidechkul BMC Public Health (2018) 18:694 Page 10 of 17 Table 7 Physical examination and medical history (Continued) Item Total Male Female χ p-value n% n% n% Tooth problem No 450 56.7 234 52.0 216 48.0 2.48 0.115 Yes 343 43.3 159 46.4 184 53.6 Headache No 557 72.1 302 54.2 275 49.4 6.55 0.010* Yes 216 27.9 91 42.1 125 57.9 Dizziness No 556 70.1 294 52.9 262 47.1 8.19 0.004* Yes 237 29.9 99 41.8 138 58.2 Peptic ulcer No 527 66.5 278 52.8 249 47.2 6.40 0.011* Yes 266 33.5 115 43.2 151 56.8 Anorexia No 707 89.2 371 52.5 336 47.5 22.18 < 0.001* Yes 86 10.8 22 25.6 64 74.4 History of injury No 713 89.9 349 48.9 364 51.1 1.05 0.305 Yes 80 10.1 44 55.0 36 45.0 History of hospital admission No 310 39.1 143 46.1 167 53.9 2.39 0.122 Yes 483 60.9 250 51.8 233 48.2 Parental history of DM No 515 64.9 262 50.9 253 49.1 1.01 0.313 Yes 278 35.1 131 47.1 147 52.9 Parental history of HT No 375 47.3 190 50.7 185 49.3 0.34 0.554 Yes 418 52.7 203 48.6 215 51.4 *Significance level at α=0.05 smoked. The participants had a high frequency of con- population, 21.9% did not know that they had HT. In sumption of dietary salt (64.4%), greasy food (62.5%), taking a closer look into tribal differences, more than sweet food (50.7%) and uncooked food (44.9%). Five fac- half of the Yao and Lisu participants had HT. This tors were found to be significantly associated with phenomenon could be attributed to the differences in T2DM: tribe, exercise, BMI, parental history of T2DM, culture and lifestyle among the hill tribe people, who and triglycerides. Another four factors were found to be consume alcohol and foods that are highly sweetened significantly associated with HT: sex, dietary salt intake, and salty and do not exercise regularly. BMI, and parental history of HT. In our study, the comorbidity rate of T2DM and HT is The results of our study revealed very interesting in- higher than that in an Indian sample in a study of Jaya formation on the prevalence of T2DM among the hill et al. [25]. However, the T2DM prevalence of our study tribe elderly populations in Thailand at 16.8%, which is sample is similar to that of a sample from a study con- 1.75 times higher than that of the Thai population [11]. ducted by Mohamed et al. [26] among the ethnic groups We also found significant differences in prevalence in northern Sudan, with a T2DM prevalence of 18.7%. among the various tribes. Meanwhile, the prevalence of Dhiraj et al. [27] reported that in different tribes of the HT was 45.5%, which is almost 1.6 times greater than population, there were different burdens of T2DM in that of the general Thai elderly population [13]. Among the sub-Himalayan region of India. This information the participants with HT in the hill tribe elderly supports the finding that the hill tribe people in Apidechkul BMC Public Health (2018) 18:694 Page 11 of 17 Table 8 Classification of participants’ characteristics by biomarkers 2 2 2 Factors Uric acid χ p-value Cholesterol χ p-value Triglyceride χ p-value Normal n (%) High n (%) Normal n (%) High n (%) Normal n (%) High n (%) Sex Male 246 (64.4) 136 (35.6) 38.63 <0.001* 286 (74.9) 96 (25.1) 14.28 <0.001* 309 (80.9) 73 (19.1) 2.44 0.118 Female 329 (83.9) 63 (16.1) 244 (67.4) 148 (32.6) 299 (76.3) 93 (23.7) Age (years) 60-69 311 (77.9) 88 (22.1) 6.04 0.049* 261 (65.4) 138 (34.6) 4.45 0.108* 303 (75.9) 96 (24.1) 6.58 0.037* 70-79 197 (71.1) 80 (28.9) 195 (78.9) 82 (21.1) 219 (79.1) 58 (20.9) ≥ 80 67 (68.4) 31 (31.6) 74 (78.7) 24 (21.3) 86 (87.8) 12 (12.2) Tribe Akha 101 (78.3) 28 (21.7) 20.19 0.018* 95 (73.6) 34 (26.4) 17.05 0.004* 99 (76.7) 30 (23.3) 8.86 0.114 Lahu 113 (85.0) 20 (15.0) 100 (75.2) 33 (24.8) 96 (72.2) 37 (27.8) Hmong 81 (64.3) 45 (35.7) 93 (73.8) 33 (26.2) 100 (79.4) 26 (20.6) Yao 96 (74.4) 33 (25.6) 82 (90.0) 47 (9.1) 98 (75.9) 31 (24.1) Karen 99 (76.7) 30 (23.3) 72 (55.8) 57 (44.2) 111 (86.0) 18 (14.0) Lisu 85 (66.4) 43 (33.6) 88 (68.8) 40 (31.2) 104 (81.3) 24 (18.7) *Significance level at α=0.05 Thailand originate from Tibet [14, 16], which is close to of T2DM, which is consistent with the findings of our those living in the sub-Himalayan region of India. study. Triglyceride levels are another factor related to Therefore, the T2DM and HT prevalence among the 6 the development of T2DM. A retrospective longitudinal hill tribes in Thailand are possibly different. large-scale study conducted between the year 2000 and A study using a mass database in Korea reported that 2012 found that every 10 mg/dL increase in triglyceride regular and frequent exercise led to reduced T2DM levels significantly increased the risk of T2DM by 4.0% mortality and morbidly rates, particularly in the elderly in the United States [37]. In addition, Ming et al. [38] re- population [28]. A study in Saudi Arabia also reported ported that an increase in triglycerides was a risk factor that sufficient physical exercise was a protective factor for type 2 diabetes among those living in rural China. against T2DM development [29]. This result is similar These studies present findings similar to those of this to the finding of our study that regular exercise and study, such that higher triglyceride levels are a risk fac- highly active physical work serve as protective factors tor for T2DM. Different tribes or races also have signifi- against T2DM among the hill tribe elderly populations cant associations with T2DM. The studies of Vitor [39] in Thailand. Regarding BMI, Kulaya et al. [30]reported and Diego et al. [40], which were conducted in the that increasing BMI was identified as a major risk fac- United States using different study designs, revealed that tor for T2DM in the Thai population. In a study of differences in the races of parents had an impact on the Asian Americans in the United States, a BMI< 23 or development of HT in their children. However, in our overweight was detected as a risk factor for T2DM de- study, there was no significant difference in HT preva- velopment [31]. Moreover, a case-control study aimed lence among the tribes. at assessing the association between BMI and T2DM in Jugal et al. [41] reported that there were several factors the Mid-Atlantic region found a heavy association be- associated with HT among those living in rural Delhi, tween increasing BMI and T2DM, after controlling for India, such as older age, alcohol use, education and chol- all confounding factors [32]. However, in a study esterol levels. However, sex was not found to be associ- among Afro-Trinidadians in the United States in 2016, ated with HT. On the other hand, Saswata et al. [42] no significant difference in BMI was found between reported that females had a greater chance of developing those who had T2DM and those who did not [33]. In HT than males in a study conducted in western India. our study, it was found that increasing BMI or over- Daily food consumption is one of the predictors for HT. weight was a risk factor for T2DM in the hill tribe Daily consumption of salty foods is one of the risk fac- elderly populations. tors of HT. This finding is supported by several studies Many studies [34–36] have reported that having a par- [43–45] that show that dietary salt intake was highly as- ental or family history of diabetes or first-degree rela- sociated with HT development in developing and devel- tives with diabetes was associated with the development oped countries and in urban and rural areas. In this Apidechkul BMC Public Health (2018) 18:694 Page 12 of 17 Table 9 Factors associated with T2DM in univariate and multivariate analyses (n = 775)** Factors T2DM OR 95%CI p-value OR 95%CI p-value adj Yes No n% n% Sex Mal 66 17.3 316 82.7 1.00 Female 64 16.3 329 83.7 0.93 1.02 -2.02 0.712 Tribe Akha 11 8.6 117 91.4 1.00 1.00 Lahu 26 19.5 107 80.5 2.58 1.37-4.85 0.013* 2.89 1.32-6.33 0.008* Hmong 11 8.1 124 91.9 0.94 0.45-1.96 0.896 0.91 0.35-2.31 0.845 Yao 26 21.5 95 78.5 2.91 1.54-5.48 0.006* 3.47 1.58-7.62 0.002* Karen 34 26.4 95 73.6 3.80 2.06-7.03 < 0.001* 5.03 2.35-10.78 < 0.001* Lisu 22 17.1 107 82.9 2.18 1.14-4.13 0.046* 2.73 1.22-6.07 0.014* Age (year) 60-69 75 18.8 324 81.2 1.00 70-79 39 14.3 234 85.7 0.72 0.50-1.02 0.127 ≥ 80 16 15.5 87 84.5 0.79 0.48-1.30 0.444 Smoking No 78 16.4 398 83.6 1.00 Ever in the past 34 23.1 113 76.9 1.53 1.04-2.24 0.064* Yes 18 11.8 134 88.2 0.68 0.43-1.08 0.177 Alcohol use No 79 15.0 447 85.0 1.00 Ever in the past 26 19.3 109 80.7 1.35 0.89-2.03 0.230 Yes 25 21.9 89 78.1 1.58 1.04 -2.42 0.072* Salty food No 151 53.5 131 46.5 1.00 Yes 266 52.1 245 47.9 0.94 0.70-1.26 0.687 Greasy food No 155 52.2 142 47.8 1.00 Yes 258 52.0 238 48.0 0.99 0.74-1.32 0.962 Sweet food No 202 51.7 189 48.3 1.00 Yes 184 45.8 218 54.2 0.78 0.59-1.04 0.097 Exercise No 21 26.6 58 73.4 1.00 1.00 Highly active physical work 45 16.7 225 83.3 0.55 0.33- 0.90 0.050* 0.48 0.25-0.91 0.024* Yes 64 15.0 362 85.0 0.48 0.30- 0.78 0.013* 0.45 0.24-0.83 0.011* BMI Normal 39 12.6 271 87.4 1.00 1.00 Underweight 13 11.4 101 88.6 0.89 0.51-1.56 0.743 0.90 0.45-1.80 0.773 Overweight 78 22.2 273 77.8 1.98 1.39- 2.82 0.001* 2.08 1.32-3.27 0.001* Parental history of DM No 217 42.1 298 57.9 1.00 1.00 Yes 149 53.6 129 46.4 1.58 1.18-2.12 0.002* 1.55 1.17-2.10 0.001* Apidechkul BMC Public Health (2018) 18:694 Page 13 of 17 Table 9 Factors associated with T2DM in univariate and multivariate analyses (n = 775)** (Continued) Factors T2DM OR 95%CI p-value OR 95%CI p-value adj Yes No n% n% Hypertension No 70 12.9 282 80.1 1.00 Yes 60 14.2 363 85.8 1.50 1.02- 2.19 0.035* Headache No 95 16.9 467 83.1 1.00 Yes 35 16.4 178 83.6 0.96 0.67-1.38 0.875 Dizziness No 86 15.9 456 84.1 1.00 Yes 44 18.9 189 81.1 1.23 0.88-1.72 0.303 Cholesterol Normal 90 17.3 430 82.7 1.00 High 38 16.1 198 83.9 0.91 0.64-1.29 0.682 Triglyceride Normal 88 14.9 504 85.1 1.00 1.00 High 40 24.4 124 75.6 1.84 1.29-2.63 0.004* 1.73 1.10-2.73 0.017* *Significance level at α=0.05 **18 participants could not provide blood specimens study, we also found that dietary salt intake among the may have occurred because they clearly did not under- hill tribe elderly populations was a significant risk factor stand the importance of laboratory interpretations. for HT development. Another factor related to HT is Moreover, most hill tribe elderly adults are not edu- BMI. Alicja et al. [46] reported that both men and cated. This finding coincides with those of studies by women had an increased risk of HT with increasing Apidechkul et al. [53]and Apidechul [54], who re- BMI, particularly among the elderly populations. A rural ported that a high proportion of the Akha elderly popu- Chinese cohort study in 2016 [47] and a study in lation and the Lahu people were in the illiterate group. Bangladesh in 2017 [48] confirmed that the increase in This finding could explain participants’ limited under- BMI had a significant association with HT development. standing of the research information and lack of These findings coincide with those of our study, which cooperation with the procedure. revealed that an increase in BMI was associated with a The researchers could not draw blood from a few partic- greater odds of HT development among the hill tribe ipants (1.26%) because of their individual peripheral vein elderly populations in Thailand. characteristics. However, nobody refused to provide infor- The study of Ghada et al. [49] in Egypt showed a mation and a specimen. Because this lack of data would strong association between a family history of HT and affect the predictive statistical model (logistic regressions), the development of HT in one’s offspring. A family his- these participants were excluded from the analysis to en- tory has been detected as a risk factor for HT among sure the accuracy of the results. Furthermore, some partic- young adults and the elderly population in several ipants had been diagnosed as T2DM and HT before countries [50–52]. starting the study, which could possibly impact the find- Some limitations have been identified in this study, ings of the study, particularly their knowledge, attitudes such as misunderstanding the NPO techniques before and practices, which are common limitations of the drawing blood specimens, language, and the inability to cross-sectional study design. Concerning this point, know- draw blood specimens in some people. Since some tar- ledge of and attitudes toward DM and HT were not in- geted hill tribe villages are located far away from the cluded in the prediction model. Moreover, if we look city, traveling to the study setting very early in the day closely, only attitude is significantly different among the to collect blood specimens was sometimes not practical. tribes. Additionally, the number of Lahu (excess of 3par- Other limitations included unclear information on the ticipants) and Hmong (excess of 10 participants) partici- research procedure and not drinking and eating food for pants exceeded the minimum requirement for the sample at least 8 hours before having blood drawn. Sometimes size due to miscommunication between the researcher there was no cooperation from the participants, which and community headman. However, these excess data did Apidechkul BMC Public Health (2018) 18:694 Page 14 of 17 Table 10 Factors associated with HT in univariate and multivariate analyses Factors HT OR 95%CI p-value OR 95%CI p-value Adj Yes No n% n% Sex Male 164 41.7 229 58.3 1.00 1.00 Female 197 49.3 203 50.7 1.35 1.02-1.79 0.034* 1.29 1.01-1.68 0.031* Tribe Akha 61 46.9 69 53.1 1.00 Lahu 61 45.9 72 54.1 0.95 0.59-1.55 0.863 Hmong 42 30.0 98 70.0 0.48 0.29-0.79 0.004* Yao 74 56.9 56 43.1 1.49 0.91-2.43 0.107 Karen 52 40.0 78 60.0 0.75 0.46-1.23 0.261 Lisu 71 54.6 59 45.4 1.36 0.83-2.21 0.215 Age (years) 60-69 173 42.2 237 57.8 1.00 70-79 134 48.0 145 52.0 1.26 0.93-1.71 0.131 ≥80 54 51.9 50 48.1 1.48 0.96-2.27 0.075 Smoking No 233 47.9 253 52.1 1.00 Ever in the past 65 43.0 86 57.0 0.82 0.56-1.18 0.293 Yes 63 40.4 93 59.6 0.73 0.51-1.06 0.100 Alcohol use No 249 46.3 289 53.7 1.00 Ever in the past 64 46.0 75 54.0 0.99 0.68-1.44 0.960 Yes 48 41.4 68 58.6 0.81 0.54-1.23 0.337 Salty food No 138 48.9 144 51.1 1.00 1.00 Yes 307 60.1 204 39.9 1.57 1.17-2.01 0.002* 1.48 1.14-2.00 0.001* Greasy food No 136 45.8 161 54.2 1.00 Yes 241 48.6 255 51.4 1.11 0.83-1.49 0.582 Sweet food No 202 51.7 189 48.3 1.00 Yes 197 49.0 205 51.0 0.89 0.68-1.18 0.454 Regular Exercise Yes 36 45.0 44 55.0 1.00 Highly active physical work 113 40.5 166 59.5 0.83 0.50-1.37 0.472 No 212 48.8 222 51.2 1.16 0.72-1.88 0.527 BMI Normal 112 35.0 208 65.0 1.00 1.00 Underweight 42 36.2 74 63.8 1.05 0.67-1.64 0.816 2.56 0.70 – 1.70 0.696 Overweight 207 58.0 150 42.0 2.56 1.87- 3.49 < 0.001* 1.37 1.01 – 1.90 < 0.001* Parental history of HT No 155 41.3 220 58.7 1.00 1.00 Yes 302 72.2 116 27.8 3.69 2.74-4.97 < 0.001* 3.38 2.81-4.48 < 0.001* Apidechkul BMC Public Health (2018) 18:694 Page 15 of 17 Table 10 Factors associated with HT in univariate and multivariate analyses (Continued) Factors HT OR 95%CI p-value OR 95%CI p-value Adj Yes No n% n% Diabetes mellitus No 282 43.7 363 56.3 1.00 Yes 70 53.8 60 46.2 1.50 1.02- 2.19 0.035* Headache No 249 43.2 328 56.8 1.00 Yes 112 51.9 104 48.1 1.41 1.03-1.94 0.029* Dizziness No 238 42.8 318 57.2 1.00 Yes 123 51.9 114 48.1 1.44 1.06-1.95 0.019* Cholesterol Normal 238 44.9 292 55.1 1.00 High 115 47.1 129 52.9 1.09 0.80-1.48 0.564 Triglyceride Normal 262 43.1 346 56.9 1.00 High 91 54.8 75 45.2 1.60 1.13- 2.26 0.007* *Significance level at α=0.05 not impact the results of study but rather supported the Acknowledgements The author would like to thank all the participants for kindly providing all power of the tests. essential information regarding the research procedures. The author is also Conducting research with the hill tribe people, grateful to all research assistants from the Center of Excellence for the Hill particularly among the elderly population, required tribe Health Research for their help in data collection. The author would like to thank The National Research Council of Thailand and Mae Fah Luang researchers to be clearly knowledgeable about the condition University, Thailand in support the grant. before reaching them. Additionally, having research assis- tants who were fluent in both Thai and the local hill tribe Funding languages was an advantage for obtaining information. This research was supported by the National Research Council of Thailand and Mae Fah Lung University, Thailand (Grant Number 77-2015). Conclusions The hill tribe elderly populations in Thailand are living Availability of data and materials The raw data supporting these findings can be found in the Additional file 1. with a high burden of T2DM and HT. T2DM and HT screening programs in these populations should be imple- mented regularly to detect early-stage and new cases. Authors’ contributions TA sought funding, designed the study protocols and procedures, collected There is an urgent need to develop proper health behavior data, analyzed and interpreted data, drafted, revised, and approved the final change models to reduce BMI and the consumption of version of the manuscript. dietary salt and greasy foods among the elderly popula- tions. Moreover, a program to encourage physical exercise Ethics approval and consent to participate is also necessary. Otherwise, Thailand must budget large Consent to participate, all study instruments and procedures were approved by amounts of money to provide care and treatment for these the Ethics Committee for Human Research, Mae Fah Laung University, Chiang Rai, Thailand (No. REH-58087). All participants received an oral and written populations in the near future. explanation and provided their consent before a voluntary agreement was witnessed and documented by signature or fingerprint. Additional file Competing interests Additional file 1: Hill tirb Elderly Data. (XLSX 545 kb) The author declares that he has no competing interests. Abbreviations BMI: Body mass index; DALYS: Disability adjusted life year; HT: Hypertension; Publisher’sNote ID: Identification; IOC: Item objective congruence; NPO: Nothing per oral; Springer Nature remains neutral with regard to jurisdictional claims in T2DM:: Type 2 diabetes mellitus; WHO: World Health Organization published maps and institutional affiliations. Apidechkul BMC Public Health (2018) 18:694 Page 16 of 17 Received: 1 February 2018 Accepted: 24 May 2018 24. American Association of Clinical Endocrinologists and American College of Endocrinology. Guidelines for management of dyslipdemia and prevention of cardiovascular disease. 2017. https://www.aace.com/files/ lipid-guidelines.pdf 25. Jaya PT, Thakur JS, Gursimer J, Sanjay J. Prevalence and determinants of References comorbidity diabetes and hypertension: evidence from non communicable 1. World Health Organization (WHO). Country and region data on diabetes disease risk factor STEPS survey, India. Diabetes & Metabolic Syndrome: and hypertension, 2017. 2017. http://www.who.int/diabetes/facts/world_ Clinical Research and Reviews. 2017;11(Sup1):S459–65. figures/en/index2.html 26. Mohamed AE, Abubakr BM, Hind EE, Kamal Y, Sufian KN, Waldie ME, et al. 2. World Health Organization (WHO). Fact sheet: Diabetes mellitus November Increasing prevalence of type 2 diabetes mellitus and impact of ethnicity in 2017. 2017. http://www.who.int/mediacentre/factsheets/fs312/en/ north Sudan. Diabetes Research and Clinical Practice. 2018;136:93–9. 3. Centers for Disease Control and Prevention (CDC). High blood pressure risk 27. Dhiraj K, Ashok KB, Dinesh K, Sunil KR. Prevalence of diabetes mellitus and factors 2017. 2017. https://www.cdc.gov/bloodpressure/risk_factors.htm its risk factors among permanently settled tribal individuals in tribal and 4. Sinna PU, Anil M, Elisabeth S, Finn BL, Reimar WT. Lifestyle factors associated urban areas in northern state of Sub-Himalayan Region of India. with type 2 diabetes and use of different glucose-lowering drugs: Cross- International of Chronic Diseases. 2014; Article ID 380597, 9 pages. https:// sectional study. PLOS ONE. 2014;9(11):e111849. https://doi.org/10.1371/ doi.org/10.1155/2014/380597. journal.pone.0111849 28. Woo YS, Taehee L, Da-Hye J, Hyeon CK. Diabetes, frequency of exercise, and 5. Funda DY, Zeynel AO. Treatment of type 2 diabetes mellitus in the elderly. mortality over 12 years: Analysis of the National Health Insurance Service- World J Diabetes. 2017;8(6):278–85. Health Screening (NHIS-HEALS) Database. J Korean Med Sci. 2018;33(8):e60. 6. Peng L, Jin-Gang Z. Blood pressure targets in the hypertensive elderly. Chin 29. Alzaheb RA, Altemani AH. The prevalence and determinants of poor Med J. 2017;130(16):1968–72. glycemic control among adults with type 2 diabetes mellitus in Saudi 7. Davies A, Janet OI, Adewale VA, Ngozi A, Emmanuel OA, Asa A, Gbolahan O. Arabia. Diabetes Metab Syndr Obes. 2018;11:15–21. Estimating the prevalence, hospitalization and mortality from type 2 30. Kulaya N, Natkamol C, Panuwat P, Jariya H. Screening high-risk populations diabetes mellitus in Nigeria: a systematic review and meta analysis. BMJ for hypertension and type 2 diabetes among Thais. WHO South-East Asia Open. 2017;7:e015424. https://doi.org/10.1136/bmjopen-2016-015424. Journal of Public Health. 2012;1(3):320–9. 8. Yashikin AP, Kravchenko J, Yashin AI, Sloan F. Mortality and macrovascular 31. Karen MK, Keith TC, Esme FT. Diabetes among Asian Americans with BMI risk in elderly with hypertension and diabetes: effect of intensive drug less than or equal to 23. Diabetes& Metabolic Syndrome: Clinical Research & therapy. American Journal of Hypertension. 2018;31(2):220–7. Reviews. 2017; https://doi.org/10.1016/j.dsx.2017.12.011 9. Sophia DF, Sunny DF. Economic burden of diabetes mellitus and its socio- 32. Michael LG, Neil W, Qian L, Veronica A, Jakob L, Mette H. The association of economic impact on household expenditure in an urban slum area. Int J body mass index with the risk of type 2 diabetes: a case-control study nested Res Med Sci. 2017 May;5(5):1808–13. in an electronic health records system in the United States. BMC Diabetology 10. World Health Organization (WHO). Diabetes country profiles, 2026: Thailand & Metabolic Syndrome. 2014;6:50. https://doi.org/10.1186/1758-5996-6-50 diabetes. 2016. http://www.who.int/diabetes/country-profiles/tha_en.pdf. 33. Shivananda N, Kheleel S, Shivang S, Vashista S, Rukaiya S, Safayah S, et al. 11. Chaicharn D, Alessandra F. Diabetes management in Thailand: a literature Investing the link between benign prostatic hypertrophy, BMI, and type 2 review of the burden, cost, and outcomes. BMC Global Health. 2013;9(11) diabetes mellitus. Diabetes& Metabolic Syndrome: Clinical Research Reviews. https://doi.org/10.1186/1744-8603-9-11. 2017;11(Siup2):S627–30. 12. World Health Organization (WHO). Raised blood pressure: Situation and 34. Deepark KD, Arun KC, Mamta S, Swayamsidha M, Binoy KM, Anoj KB. Effect trend. 2017. http://www.who.int/gho/ncd/risk_factors/blood_pressure_ of parental history of diabetes on markers of inflammation insulin resistance prevalence_text/en/. and atherosclerosis in first degree relatives of patients with type 2 diabetes 13. World Health Organization (WHO)-Thailand. Technical consultation to mellitus. Diabetes & Metabolic Syndrome: Clinical Research & Reviews. 2017; strengthen hypertension care in Thailand, 2017. 2017. http://www.searo. https://doi.org/10.1016/j.dsx.2017.12.004 who.int/thailand/news/htn-meeting-report-draft24nov.pdf?ua=1 35. Kara MG, Shannon KC, Duck-chul L, Xuemei SS, Steven PH, Steven NB. 14. Princess Maha Chakri Siridhorn Anthropology center. Hill tribe. 2018. Association of cardiorespiratory fitness and parental history of diabetes with risk http://www.sac.or.th/main/index.php of type 2 diabetes. Diabetes research and Clinical Practice. 2012;95(3):425–31. 15. The hill tribe welfare and development center, Chiang Rai province. Hill 36. Chaochen W, Hiroshi Y, Koji T, Hideaki T, Keiko W, Yuanying L, et al. tribe population. The hill tribe welfare and development center. Chiang Rai: Association between parental history of diabetes and the incidence of type Ministry of Interior; 2017. p. 22–31. 2 diabetes mellitus differs according to the sex of the parent and offspring’s 16. Tawatchai A. A 20-year retrospective cohort study of TB infection among body weight: A finding from a Japanese worksite based cohort study. the hill tribe HIV/AIDS populations, Thailand. BMC Infectious Disease. 2016; Preventive Medicine. 2015;81:49–53. 16:72. https://doi.org/10.1186/s12879-016-1407-4. 37. Aman B, Eytan C, Elad G, Pearl L, Moshe G, Ilan K. Triglyceride levels and risk 17. Tawatchai A, Onn L, Sirinan S. Inequity in accessing health care service in of type 2 diabetes mellitus: A longitudinal large study. J Investig Med. 2016; Thailand in 2015: A case study of the hill tribe people in Mae Fah Luang 64(2):383–7. district, Chiang Rai, Thailand. J Health Res. 2016;30(1):67–71. 38. Ming Z, Bingyuan W, Yu L, Xizhuo S, Xinping L, Chongjian W, et al. 18. Sudkhed D, Tawatchai A, Rachanee S, Amornrat A, Onnalin S, Bukhari P, Cumulative increased risk of incident type 2 diabetes mellitus with et al. Patterns and perception of alcohol drinking among the Lahu people, increasing triglyceride glucose index in normal-weight people: the rural northern Thailand. J Health Res. 2018;32(3):67–71. Chinese cohort. BMC Cardiovascular Diabetology. 2017;16(30): https://doi. 19. Incidence and risk factors for type 2 diabetes mellitus in transitional org/10.1186/s12933-017-0514-x Thailand: results from the Thai cohort study. BMJ Open. 2016;6:e014102. https://doi.org/10.1136/bmjopen-2016-014102 39. Victor G. Pregravid hypertension may have different secondary sex ratio 20. World Health Organization (WHO). Definition and diagnosis of diabetes effects in different races in the United States. International Journal of mellitus and intermediate hyperglycemia. 2017. http://www.who.int/ Cardiology. 2017;249:333–5. diabetes/publications/Definition%20and%20diagnosis%20of%20diabetes_ 40. Diego GD, Arthur EM, Raphael MR, Romulo AF, Bruna TC, Mariana RP, et new.pdf al. Association between hypertension in adolescents and the health risk factors of their parents: an epidemiological family study. Journal of 21. The hill tribe welfare and development center, Chiang Rai province. Hill American Society of Hypertension. 2018; https://doi.org/10.1016/j.jash. tribe population. In: The hill tribe welfare and development center. Chiang 2017.12.011 Rai: Ministry of Interior; 2015. p. 16–27. 41. Jugal K, Neeru G, Charu K, Neeta K. Prevalence of hypertension and 22. American College of Cardiolog. 2017 Guideline for prevention, determination of its risk factors in rural Delhi. International Journal of detection, evaluation, and management of high blood pressure in Hypertension. 2016; https://doi.org/10.1155/2016/7962595. adult. file:///Users/macbookpro13/Downloads/Guidelines_Made_Simple_ 2017_HBP.pdf 42. Saswata G, Simantini M, Anamitra B. Sex difference in the risk profile of 23. World Health Organization (WHO). BMI classification in Asian population. hypertension: A cross-sectional study. BMJ open. 2016:e010085. https://doi. 2015. http://apps.who.int/bmi/index.jsp?introPage=intro_3.html org/10.1136/bmjopen-2015-010085. Apidechkul BMC Public Health (2018) 18:694 Page 17 of 17 43. Sung KH. Dietary salt intake and hypertension. Electrolyte Blood Press. 2014; 12(1):7–18. 44. William BF, David GE, Claudine TJ, William SW. Dietary sodium and health: more than just blood pressure. J Am Coll Cardiol. 2015;65(10):1042–50. 45. Mandapaka RT, Rachabathuni S. Prevalence of hypertension and its relationship between dietary salt intake in urban population. J Community Med Health. 6:426. https://doi.org/10.4172/2161-0711.1000426. 46. Alicja S, Anna L. BMI, hypertension and low bone mineral density in adult men and women. HOMO-Journal of Comparatie Human Biology. 2012;63(4):282–91. 47. Ming Z, Yang Z, Haohang S, Xinping L, Chongjian W, Linlin L. Effect of dynamic change in body mass index on the risk of hypertension: results from the rural Chinese chohort study. International Journal of Cardiology. 2017;238:117–22. 48. Mohammed MR, Shamima A, Jenny J, Mohammed SR, Papia S. Trend, projection, and appropriate body mass index cut-off point for diabetes and hypertension in Bangladesh. Diabetes Research and Clinical Practice. 2017; 126:43–53. 49. Ghada Y, Ibrahim ET, Dina O, Ahmed S, Essam B, Zeinab A, et al. Familial history of hypertension as a predictor of increased arterial stiffness in normotensive offspring. The Egypt Heart Journal. 2017;69(1):37–44. 50. Menghetti E, Strisciuglio P, Spagnolo A, Carletti M, Paciotti G, Muzzi G, et al. Hypertension and obesity in Italian school children: The role of diet, life style and family history. Nutrition, metabolism and Cardiovascular Diseases. 2015;25(6):602–7. 51. Dorairaj P, Panniyammakal J, Shreeparna G, Roopa S, Vamadevan SA, Dimple K, et al. Prevalence and incidence of hypertension: results from a representative cohort of over 16,000 adults in three cities of South Asia. Indian Heart Journal. 2017;69(4):434–41. 52. Mercedes RC, Guadalupe XA, Shrikant IB. Association of cardiovascular risk factors between Hispanic/Latino parents and youth: the Hispanic community health study/study of Latino youth. Annals of Epidemiology. 2017;27(4):260–8. 53. Tawatchai A, Pilasinee W, Siriyaporn S, Thapakorn R. Health situation of Akha hill tribe in Chiang Rai province, Thailand. Journal of Public Health and Development. 2016;14(1):77–97. 54. Tawatchai A. Prevalence of thalassemia carriers among the Lahu hill tribe population, Chiang Rai, Thailand. Asian Biomedicine. 2015;9(4):527–33. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Public Health Springer Journals

Prevalence and factors associated with type 2 diabetes mellitus and hypertension among the hill tribe elderly populations in northern Thailand

BMC Public Health , Volume 18 (1) – Jun 5, 2018
Free
17 pages

Loading next page...
 
/lp/springer_journal/prevalence-and-factors-associated-with-type-2-diabetes-mellitus-and-5n0b94ca4h
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Medicine & Public Health; Public Health; Medicine/Public Health, general; Epidemiology; Environmental Health; Biostatistics; Vaccine
eISSN
1471-2458
D.O.I.
10.1186/s12889-018-5607-2
Publisher site
See Article on Publisher Site

Abstract

Background: Type 2 diabetes mellitus (T2DM) and hypertension (HT) are major noncommunicable health problems in both developing and developed countries, including Thailand. Each year, a large amount of money is budgeted for treatment and care. Hill tribe people are a marginalized population in Thailand, and members of its elderly population are vulnerable to health problems due to language barriers, lifestyles, and daily dietary intake. Methods: An analytic cross-sectional study was conducted to estimate the prevalence of T2DM and HT and to assess the factors associated with T2DM and HT. The study populations were hill tribe elderly adults aged ≥ 60 years living in Chiang Rai Province, Thailand. A simple random method was used to select the targeted hill tribe villages and participants into the study. A validated questionnaire, physical examination form, and 5-mL blood specimen were used as research instruments. Fasting plasma glucose and blood pressure were examined and used as outcome measurements. Chi-square tests and logistic regression were used for detecting the associations between variables at the significance level alpha=0.05. Results: In total, 793 participants participated in the study; 49.6% were male, and 51.7% were aged 60-69 years. A total of 71.5% were Buddhist, 93.8% were uneducated, 62.9% were unemployed, and 89 % earned an income of < 5,000 baht/month. The overall prevalence of T2DM and HT was 16.8% and 45.5%, respectively. Approximately 9.0% individuals had comorbidity of T2DM and HT. Members of the Lahu, Yao, Karen, and Lisu tribes had a greater odds of developing T2DM than did those of the Akha tribe. Being overweight, having a parental history of T2DM, and having high cholesterol were associated with T2DM development. In contrast, those who engaged in highly physical activities and exercise had lower odds of developing T2DM than did those who did not. Regarding HT, being female, having a high dietary salt intake, being overweight, and having a parental history of HT were associated with HT development among the hill tribe elderly populations. Conclusions: The prevalence of T2DH and HT among the hill tribe elderly populations is higher than that among the general Thai population. Public health interventions should focus on encouraging physical activity and reducing personal weight, dietary salt intake, and greasy food consumption among the hill tribe elderly. Keywords: Type 2 diabetes mellitus, Hypertension, Hill tribe, Elderly, Thailand Correspondence: tk2016ms@gmail.com; tawatchai.api@mfu.ac.th Center of Excellence for the Hill tribe Health Research, Mae Fah Luang University, Chiang Rai, Thailand School of Health Science, Mae Fah Luang University, Chiang Rai, Thailand © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Apidechkul BMC Public Health (2018) 18:694 Page 2 of 17 Background the number of deaths was 21,830 cases (9,430 males, Type 2 diabetes mellitus (T2DM) and hypertension 12,400 females) [10]. The average cost of each T2DM case (HT) are common noncommunicable diseases among in attending hospital services per year was 598US$ for an elderly adults aged ≥ 60 years in both developing and de- independent case and 2,700US$ for a disabled case. veloped countries [1]. The prevalence of T2DM and HT Therefore, Thailand spends a large amount of money on varies according to age, sex, and race [2, 3]. There are the health care system annually [11]. different factors associated with T2DM and HT in differ- High blood pressure is a key risk factor for many ent populations, particularly among those with different diseases, including heart attack and stroke. In 2017, lifestyles and cultures [3, 4]. Older populations are the WHO estimated that more than one billion people most vulnerable to the development of T2DM and HT had HT caused 12.8% of all deaths and accounted for [5, 6]. T2DM and HT have become major causes of 57 million disability-adjusted life years (DALYs) or a morbidity and mortality of elderly populations in all total of 3.7% DALYs every year [12]. Thailand re- countries [7, 8]. The impact of T2DM and HT is not ported that 29.0% of adult Thais had HT, and only limited to physical and mental consequences; rather, it 37.0% for those people who had been diagnosed had also affects family and national economics [9]. Health their blood pressure under control in 2017 [13]. The professionals in health care institutes must manage the number of resistant HT patients in all health insti- maintenance of plasma glucose levels among T2DM pa- tutes in the entire country has increased from tients and blood pressure among HT patients using dif- 3,946,902 cases in 2013 to 5,584,007 cases in 2017 ferent regiments of drugs for their entire lives. With [13]. The statistics represent the full picture of the these demands, there are required numbers of health situation in Thailand, but there is no information professionals and large amount of financial input needed available on any specific subgroup of populations, to operate the treatment and care system each year. such as the hill tribe population. Patients need to frequently attend a clinic to meet and The hill tribe people are those who have migrated receive care from a doctor. Otherwise, many complica- from the southern region of China to Thailand in the tions could possibly develop, resulting in intensive and past century [14]. They are divided into six different complicated methods of treatment and care. main groups: Akha, Lahu, Karen, Hmong, Yao, and In 2014, the WHO estimated that 422 million people Lisu [15]. Approximately 2.5 million of the hill tribe worldwide were suffering from T2DM, which accounted people were living in Thailand in 2017 [16]. They for 8.5% of the prevalence among people over 18 years have their own culture, language and lifestyles, par- old. The prevalence is increasing among people aged > 30 ticularly in daily cooking. Some tribes use a high vol- years old, particularly in low- and middle-income coun- ume of oil for cooking, whereas other tribes use a tries. People aged ≥ 60 years old are also commonly de- high volume of salt for their daily food [14, 17]. Most fined as a vulnerable population for T2DM [2]. of them have similar cultural patterns in terms of Commonly, T2DM is a disease that progresses slowly using alcohol, particularly for religious rituals [18]. from its onset, and it may be diagnosed several years later. In 2017, the hill tribe elderly populations lived ac- T2DM is a major cause of other health problems, such as cording to their own traditional lifestyle and living blindness, kidney failure, heart attacks, stroke, and lower environment. They consumed drinks and foods pre- limb amputation. The WHO also reported that 1.6 million pared traditionally. Individual health care was mainly deaths were directly caused by diabetes, and almost half of based on their local healing patterns. With the prob- all deaths attributable to high blood glucose occurred be- lems of distance, language and discrimination, their fore the age of 70 years [2]. This finding reflects the need access to the Thai health care system was poor [19]. to regularly investigate those vulnerable to an early diag- Therefore, access to modern medical care is not nosis and determine ways of obtaining a better prognosis. common, especially for those who live very far from In 2016, the total T2DM prevalence among the Thai the city. Ultimately, the findings of the study could population was 9.6%: 9.1% in males and 10.1% in females. support the development of the health care service The total number of deaths caused by T2DM was 20,570 system for the hill tribe elderly populations. The cases; in the 30-69 year age group, the number of findings could also be used for the development of deaths was 8,120 cases (3,610 males, 4,510 females) and DM and HT prevention and control measures in in the ≥ 70 years age group, the number of deaths was these populations. Currently, there is no available in- 12,450 cases (4,760 males, 7,690 females). Moreover, formation about T2DM and HT among these popula- total number of deaths attributable to high blood tion groups. Therefore, the study aimed to estimate glucose was 35,640 cases; in the 30-69 year age the prevalence and factors associated with DM and group, the number of deaths was 13,810 cases (7,220 HT among the hill tribe elderly populations in north- males, 6,590 females) and in the ≥ 70 years age group, ern Thailand. Apidechkul BMC Public Health (2018) 18:694 Page 3 of 17 Methods Sample selection and preparing the participants Study design and participants The list of the hill tribe villages in Chiang Rai Province Study design was requested from the Hill Tribe Welfare and Develop- A cross-sectional study was conducted to gather infor- ment Center in Chiang Rai [21]. There were 749 hill mation from the selected subjects. tribe villages in Chiang Rai, which breakdown into 316 Lahu villages, 243 Akha villages, 63 Yao villages, 56 Hmong villages, 36 Karen villages, and 35 Lisu villages. Study setting In 2016, a total of 41,366 hill tribe families lived in The study was conducted along 16 districts in Chiang Chiang Rai Province. Rai Province, which is located in Thailand. Permission to access the villages had been granted by the District Government Officer. Sixty hill tribe villages, Study population or 10 villages in each tribe, were selected by a simple The study population was comprised of hill tribe elderly random method. A village headman was contacted and adults aged ≥ 60 years old who had lived in the study informed of all essential information regarding the re- setting for at least 3 years. search objective and its protocol. The list of elderly people who met the inclusion and exclusion criteria in the village was sent to the researcher. A simple random Eligible population method was used to select 13 individuals in each village, Elderly adults with the following characteristics were eli- after which they were invited to participate in the study. gible for the study: a) being classified as a member of After informing the village headman about the research the hill tribe by verbal confirmation, b) being ≥ 60 years objectives and protocols, some tribes collected more old, c) living in the study area for 3 years at the date of than the minimum required sample size: Lahu (an excess data collection, and d) having the ability to provide es- of 3 participants) and Hmong (an excess of 10 partici- sential information. Those who had been diagnosed with pants). Those who agreed to join the project were in- type 1 diabetes mellitus, which requires daily administra- formed of all research processes, including the tion of insulin, were excluded from the study. preparation of NPO (nothing per oral) for at least 8 hours for the blood specimen collection on the next day (Fig. 1). Sample size Six research assistants fluent in Thai and in one of The sample size was calculated by Epi-Info version 7.2 the six hill tribal languages were recruited. Selected re- (US Centers for Disease Control and Prevention, At- search assistants were trained in procedures, and the lanta, GA). By setting the alpha error at 0.05, the power required documents were completed three days before at 0.8, the previous prevalence of T2DM among the working in the field. Most of the hill tribe elderly exposed group at 18.0%, and the prevalence among the populations do not speak Thai. Therefore, there was a unexposed group at 0.07% [20], the sample size was cal- need to obtain complete information using the culated to be a minimum of 705 participants. Increasing research assistants. Recruiting young adults to help as the sample size by 10.0% for error resulted in 775 partic- research assistants was possible because hill tribe ipants required. community members younger than 25 years old had Since the sample size was calculated at 775 participants, already completed secondary level education in Thai at least 130 participants were needed in each tribe. schools. 749 hill tribe villages 10 villages/tribe 13-14 persons/village 130 Akhas 133 Lahus 140 Hmongs 130 Yaos 130 Karens 130 Lisus Fig. 1 Flowchart of participants’ selection. Sixty hill tribe villages were randomly selected from 749 villages, and 13-14 elderly people in each village were recruited into the study Apidechkul BMC Public Health (2018) 18:694 Page 4 of 17 Measurement Blood pressure was assessed twice in all participants Research instruments with a 15-minute gap between assessments, and in case A questionnaire, physical examination form, a manual systolic and/or diastolic blood pressures were greater sphygmomanometer, and a 5-mL blood specimen served than 90 mmHg and/or 150 mmHg, respectively, it was nd as research instruments. A questionnaire was developed assessed again after 15 minutes of rest following the 2 from the review literature. After completion of the first assessment. Participants with 90 mmHg and/or 140 draft, the validity was detected by the item-objective mmHg of systolic and diastolic blood pressures, respect- congruence (IOC) technique in which three external ex- ively, were diagnosed as HT patients [22]. perts in relevant fields verified the validity. Questions Body mass index (BMI) was classified into three cat- with scores less than 0.5 were excluded, those with egories, according to the WHO guidelines for Asian scores 0.5-0.69 were revised, and those with scores populations: underweight (BMI≤18.5), normal weight greater than 0.7 were defined as acceptable to use. The (BMI=18.51-22.99), and overweight (BMI ≥ 23.00) [23]. questionnaire was also tested for reliability by pilot test- A 5-mL blood specimen was collected from a peripheral ing it in 15 similar participants in the Ban San Ti Suk vein puncture. After blood was drawn, a 3-mL blood spe- hill tribe village using the test-retest method. Questions cimen was collected and stored in a sodium fluoride tube with Cronbach’s alpha ≥ 0.5 were included in the form. to detect fasting plasma glucose. Another 2-mL blood spe- Ultimately, there were 28 questions in three parts in- cimen was collected and stored in a clot blood clot tube cluded in the questionnaire, which presented an overall for detecting lipid profiles. Uric acid, cholesterol, and tri- Cronbach’s alpha of 0.77. glycerides were assessed in mg/dL. Participants with uric In the first parts, 13 questions were used to collect acid ≥ 7 mg/dL, cholesterol ≥ 200 mg/dL, and triglycerides participants’ general information, such as age, sex, edu- ≥ 200 mg/dL were defined as a high-level group [24]. Par- cation, and religion. Fifteen questions were included in ticipants with fasting plasma glucose ≥ 126 mg/dL were the second part, including questions about health behav- asked to provide another blood specimen within a week to iors such as “Do you smoke?”, “Do you drink alcohol?”, determine type 2 diabetes stage. and “Do you use methamphetamine?”. In these ques- tions, the three answer choices were “Yes”, “Ever in the Procedures past”, and “No”. Data gathering procedures Several questions were asked regarding daily food con- After the consent form was obtained, a 5-mL blood spe- sumption and exercise, such as “Do you usually eat a salty cimen was collected. Participants were asked to diet?”, “Do you favor having greasy food?”,and “Do you complete the questionnaire in a private room in the vil- like to eat sweet food?” Two answer choices were provided lage with the help of the research assistants. A trained for the questions, “Yes” and “No”. However, for exercise physician examined the physical health of all participants practices, the three answer choices were “No”, “Highly ac- in a proper room. A small gift was given to participants tive physical work, such as farmer and labor”,and “Yes”. after they completed the questionnaire. Questions on medications, history of T2DM and HT, and parental history of T2DM and HT were also included. Statistical analysis To confirm the diagnosis, all participants who responded Descriptive statistics, such as the means, minimums, that they had T2DM and HT were asked to present the maximums, standard deviations, and percentages, were log-book from a hospital. In Thailand, all DM and HT used to explain the general characteristics of the partici- cases are provided individual log-books to use to collect pants. Chi-square tests and logistic regressions were medical information and make appointments. used to detect the associations between variables at the The last part consisted of twenty-one items of a phys- significance level α=0.05. Logistic regression was used to ical examination form, which is used at Mae Fah Luang detect the associations between variables in both univar- University Hospital. This form resembles a checklist and iate and multivariate models. The “ENTER” mode was can include more information if required. A manual used to select the significant variables in the model. The sphygmomanometer was used for assessing blood significance level (alpha) was set at 0.05 in both univari- pressure. ate and multivariate analyses. Variables that were found to be significant in the univariate analysis were retained Variables and measurements in the multivariate analysis. In the multivariate model, T2DM in the study was identified by the following: a) the most nonsignificant variable was deleted from the having no history of a medical diagnosis, such as type 1 model before running the second step. The model was diabetes mellitus, at a particularly early age after birth, analyzed until all remaining variables were found to be b) having shown a fasting plasma glucose ≥ 126 mg/dL significant at an alpha level of 0.05, and the results were twice on different days [20]. interpreted. Apidechkul BMC Public Health (2018) 18:694 Page 5 of 17 Table 1 General characteristics of the study participants Table 1 General characteristics of the study participants (Continued) Characteristics Number Percent Characteristics Number Percent Total 793 100.0 Merchant 11 1.4 Sex Labor 19 2.4 Male 393 49.6 Other 12 1.5 Female 400 50.4 Monthly family income (baht) Thai ID card 0 69 8.7 Yes 745 93.9 ≤5,000 707 89.2 No 48 6.1 ≥5,001 17 2.1 Tribe Debt (baht) Akha 130 16.4 0 673 84.9 Lahu 133 16.8 ≤5,000 14 1.8 Hmong 140 17.6 5,001-10,000 11 1.4 Yao 130 16.4 10,001-50,000 58 7.3 Karen 130 16.4 ≥50,001 37 4.6 Lisu 130 16.4 Age (years) 60-69 410 51.7 Results 70-79 279 35.2 Characteristics of participants ≥ 80 104 13.1 In total, 793 participants were recruited into the study. Proportions of participants were mostly equal by sex and Religion among the six tribes. A few people had no Thai identifica- Buddhism 567 71.5 tion card (6.1%), with an equal proportion among the Christianity 225 28.4 tribes. The majority were aged 60-69 years (51.7%), with Islam 1 0.1 an average age of 70.1 years (range=60-100, SD=7.57, Education max=100, and min=60). The majority of the sample prac- None 739 93.8 ticed Buddhism (71.5%) and had no education (94.8%). A few people lived alone (6.1%), and most participants were Primary School 41 5.2 married (66.8%). Regarding economic status, 89.2% had an High School 8 1.0 income of ≤ 5,000 baht/month (mean=1,129 baht, Resides with SD=1,273), and 84.9% had no debt (Table 1). Child 559 70.5 There were no statistical differences in the distribution Cousin 12 1.5 of participants according to sex and tribe in three differ- Spouse 174 21.9 ent age categories (60-69, 70-79, and ≥ 80 years). A few of the hill tribe elderly adults had the ability to commu- Alone 48 6.1 nicate in Thai: 19.5% could speak, 19.5% could under- Marital status stand, 2.0% could read, and 1.6% could write fluently. Single 15 1.9 Males had significantly better Thai communication skills Married 524 66.8 than females in all four domains: speaking, understand- Divorced 20 2.5 ing, reading, and writing. Widow 226 28.8 The prevalence of T2DM and HT was 16.8% and 45.5%, respectively. Seventy-five participants had been Number of family member (persons) diagnosed with T2DM before being recruited into the 1 40 5.0 study. Among these participants, 8 (10.6%) had high 2 116 14.6 fasting glucose or were unable to control blood glucose 3-5 301 38.0 after medication. Fifty-five participants (7.7%) were de- 6 336 42.4 tected as new T2DM cases (Table 2). However, 18 par- Occupation ticipants (1.2%) could not draw blood specimens. Two hundred and forty participants (30.3%) had been Unemployed (retired) 499 62.9 diagnosed with HT, among whom 37.9% were unable to Farmer 252 31.8 control their blood pressure after medication. After Apidechkul BMC Public Health (2018) 18:694 Page 6 of 17 Table 2 Prevalence of T2DM and HT among the participants Table 3 Comparison of T2DM and HT by participants’ characteristics Chracteristics Number Percent 2 2 Characteristic T2DM χ p-value HT χ p-value Medical history of T2DM Yes No Yes No No 718 90.5 (%) (%) (%) (%) Yes 75 9.5 Sex Effective control of blood glucose by daily medication Male 66 316 0.13 0.712 164 229 4.52 0.034* (17.3) (82.7) (41.7) (58.3) No 8 10.6 Female 64 329 197 203 Yes 67 89.4 (16.3) (83.7) (49.3) (50.7) Fasting plasma glucose level among non-DM diagnosed Age (years) Normal 645 89.8 60-69 75 324 2.49 0.287 173 237 4.25 0.119 High (T2DM) 55 7.7 (18.8) (81.2) (42.2) (57.8) 70-79 39 234 134 145 (Missing=18, 2.5%) (14.3) (85.7) (48.0) (52.0) Prevalence of T2DM=16.8% ≥80 16 87 54 50 Medical history of HT (15.5) (84.5) (51.9) (48.1) No 553 69.7 Tribe Yes 240 30.3 Akha 11 117 24.48 <0.001* 61 69 26.45 <0.001* (8.6) (91.4) (46.9) (53.1) Effective control of blood pressure by daily medication Lahu 26 107 61 72 No 91 37.9 (19.5) (80.5) (45.9) (54.1) Yes 149 62.1 Hmong 11 124 42 98 (8.1) (91.9) (30.0) (70.0) Blood pressure level among non-HT diagnosed Yao 26 95 74 56 Normal 432 78.1 (21.5) (78.5) (56.9) (43.1) High (HT) 121 21.9 Karen 34 95 52 78 Prevalence of HT=45.5% (26.4) (73.6) (40.0) (60.0) Having both T2DM and HT 70 9.0 Lisu 22 107 71 59 (17.1) (82.9) (54.6) (45.4) The overall prevalence of T2DM among the participants The overall prevalence of HT among the participants *Significance level at α=0.05 those who had no history of HT diagnosis and medication With regard to the physical health and medical his- were seen, 121 participants (21.9%) were detected as new tory among the participants, 45.0% were overweight, HT cases. Finally, 70 cases (9.0%) were determined to have 6.8% were disabled persons, 15.0% had sleeping prob- both T2DM and HT: 36 males and 34 females (Table 2). lems, 9.7% had cataracts, 28.7% had hearing problems, There was statistical significance in the proportion of and 43.3% had tooth problems (Table 7). participants with T2DM and HT by sex and tribe. Only There were statistically significant differences in the qual- the participants with T2DM showed a statistically signifi- ity of uric acid and cholesterol according to sex, age cat- cant difference in proportion (Table 3). egory, and tribe. A greater proportion of males, individuals Health behaviors among the participants indicated that in higher age categories, and Lahu and Lisu tribe members 19.7% smoked, 14.6% drank alcohol, 44.9% ate uncooked hadhighuricacidlevelsthandid females, thoseinyounger food, 23.8% chewed tobacco, and 10.1% did not exercise age categories, and members of other tribes. Only age cat- regularly. A comparison of health behaviors such as smok- egory and tribe showed significant differences on the level ing, alcohol use, eating uncooked food, and regular exercise of triglycerides; a greater proportion of those in lower age among the tribes showed statistically significant differences categories had high cholesterol than those in higher age (Table 4). Additionally, there were significant sex differ- categories. A greater proportion of members of the Lahu ences in the following health behaviors: smoking; alcohol and Akha tribes were in the high cholesterol group com- use; the consumption of uncooked food, salty food, greasy pared to those in the remaining tribes (Table 8). food, and sweet food; opium use; chewing tobacco; and In themultivariatemodel, fivefactors were associated regular exercise (Table 5). with T2DM: tribe, exercise, BMI, parental history of Most participants had moderate levels of T2DM, and triglycerides. The Lahu, Yao, Karen, and health-related knowledge, attitudes, and practices. Lisu tribes had greater odds of developing T2DM than Only the distribution of attitudes by tribe showed the Akha tribe, with OR =2.89 (95%CI=1.32-6.33), adj statistical significance (Table 6). OR =3.47 (95%CI=1.58-7.62), OR =5.03 (95%CI= adj adj Apidechkul BMC Public Health (2018) 18:694 Page 7 of 17 Table 4 Characteristics of health behaviors by tribe Health behaviors Tribe χ p-value Total Akha Lahu Hmong Yao Karen Lisu n% n% n % n % n % n% n% Smoking No 486 61.3 94 19.3 70 14.4 106 21.8 72 14.8 48 9.9 96 19.8 79.02 < 0.001* Ever in the past 151 19.0 12 7.9 33 21.9 11 7.3 29 19.2 50 33.1 16 10.6 Yes 156 19.7 24 15.4 30 19.2 23 14.7 29 18.6 32 20.5 18 11.5 Alcohol use No 538 67.8 99 18.4 92 17.1 109 20.3 88 16.4 77 14.3 73 13.6 43.93 < 0.001* Ever 139 17.5 13 9.4 29 20.9 14 10.1 17 12.2 25 18.0 41 29.5 Yes 116 14.6 18 15.5 12 10.3 17 14.7 25 21.6 28 24.1 16 13.8 Methamphetamine use No 776 97.9 124 16.0 132 17.0 137 17.7 126 16.4 128 16.5 129 16.6 12.15 0.275 Ever in the past 2 0.3 0 0.0 0 0.0 0 0.0 1 50.0 1 50.0 0.0 0.0 Yes 15 1.9 6 40.0 1 6.7 3 20.0 3 20.0 1 6.7 1 6.7 Opium use No 723 91.2 112 15.5 125 17.3 124 17.2 115 15.9 123 17.0 124 17.2 15.77 0.106 Ever in the past 54 6.8 12 22.2 6 11.1 12 22.2 12 22.2 7 13.0 5 9.3 Yes 16 2.0 6 37.5 2 12.5 4 25.0 3 18.8 0 0.0 1 6.3 Eating uncooked food No 385 48.5 79 20.5 74 19.2 68 17.7 69 17.9 43 11.2 52 13.5 29.65 < 0.001* Ever in the past 52 6.6 5 9.6 6 11.5 9 17.3 8 15.4 11 21.2 13 25.0 Yes 356 44.9 46 12.9 53 14.9 63 17.7 53 14.9 76 21.3 65 18.3 Chewing No 604 76.2 70 11.6 108 17.9 135 22.4 128 21.2 100 16.6 63 10.4 159.80 < 0.001* Yes 189 23.8 60 31.7 25 13.2 5 2.6 2 1.1 30 15.9 67 35.4 Regular exercise No 80 10.1 22 27.5 7 8.8 15 18.8 7 8.8 22 27.5 7 8.8 37.50 < 0.001* Yes 433 54.6 68 15.7 88 20.3 75 17.3 66 15.2 54 12.5 82 18.9 Highly active physical work 280 35.3 40 14.3 38 13.6 50 17.9 57 20.4 54 19.3 41 14.6 *Significance level at α=0.05 2.35-10.78), and OR =2.73 (95%CI=1.22-6.07) respectively. dietary salt intake, BMI, and parental history of HT. adj Those who were overweight had greater odds of developing Females had greater odds of developing HT than males, T2DM than those with normal weight, with OR =2.08 with OR =1.29 (95%CI=1.01-1.68). Those who had adj adj (95%CI=1.32-3.27). Those who had a parental history of dietary salt intake had greater odds of developing HT T2DM had greater odds of developing T2DM than those than those who did not, with OR =1.48 adj who did not, with OR =1.55 (95%CI=1.17-2.10). Those (95%CI=1.14-2.00). Those who were overweight had adj with high cholesterol had greater odds of developing greater odds of developing HT than those with normal T2DM than those with low cholesterol, with OR =1.73 weight, with OR =1.37 (95%CI=1.01-1.90), and those adj adj (95%CI=1.10-2.73). Those who engaged in high levels of who had a parental history of HT had greater odds of physical activity and exercise had lower odds of developing developing HT than those who did not, with OR =3.38 adj T2DM than those who did not, with OR =0.48 (95%CI=2.81-4.48) (Table 10). adj (95%CI=0.25-0.91) and OR =0.45 (95%CI=0.24-0.83), re- adj spectively (Table 9). Discussion Four factors were found to be associated with HT after Members of the hill tribe elderly population are living controlling for all possible confounding variables: sex, with a high burden of T2DM and HT in Thailand. There Apidechkul BMC Public Health (2018) 18:694 Page 8 of 17 Table 5 Comparison of health behavior by sex Health behvaior Total Male Female χ p-value n % n% n% Smoking No 486 61.3 151 31.1 335 68.9 173.52 < 0.001* Ever in the past 151 19.0 125 82.8 26 17.2 Yes 156 19.7 117 75.0 39 25.0 Alcohol use No 538 67.8 169 31.4 369 68.6 222.02 < 0.001* Ever in the past 139 17.5 117 84.2 22 15.8 Yes 116 14.6 107 92.2 9 7.8 Consumption of uncooked food No 385 48.5 106 27.5 279 72.5 145.24 < 0.001* Ever in the past 52 6.6 37 71.2 15 28.8 Yes 356 44.9 250 70.2 106 29.8 Salty food No 282 35.6 106 37.6 176 62.4 25.05 < 0.001* Yes 511 64.4 287 56.2 224 43.8 Greasy food No 297 37.5 194 65.3 103 34.7 47.12 < 0.001* Yes 496 62.5 199 40.1 297 59.9 Sweet food No 391 49.3 216 55.2 175 44.8 9.96 0.0016* Yes 402 50.7 177 44.0 225 56.0 Opium use No 723 91.2 339 46.9 384 53.1 23.95 < 0.001* Ever in the past 54 6.8 43 79.6 11 20.4 Yes 16 2.0 11 68.8 5 31.3 Methamphetamine use No 776 97.9 381 49.1 395 50.9 3.69 0.079 Yes 17 2.1 12 70.6 5 29.4 Chewing No 604 76.2 313 51.8 291 48.2 5.19 0.023* Yes 189 23.8 80 42.3 109 57.7 Regular exercise No 433 54.6 184 42.5 249 57.5 26.05 < 0.001* Highly active physical work 280 35.3 173 61.8 107 38.2 Yes 80 10.1 36 45.0 44 55.0 *Significance level at α=0.05 are several factors associated with HT and T2DM, such read and write in Thai. The prevalence of T2DM and as behaviors related to daily living, culture and food HT was 16.8% and 45.5%, respectively, of which 7.7% practices. Most members of the hill tribe elderly popula- and 21.9% represented the incident rates for T2DM and tion have no education and low economic status. Very HT, respectively. Moreover, 9.3% of T2DM participants few have Thai ID cards, which is usually used to access and 37.9% of HT participants could not control their all public services in Thailand, including health care ser- plasma glucose and blood pressure after having daily vices [17]. Only one-fourth of the participants were able medication. The comorbidity rate was approximately to speak and understand Thai, and a few people could one-fourth of the participants who used alcohol and Apidechkul BMC Public Health (2018) 18:694 Page 9 of 17 Table 6 Comparison on knowledge, attitudes, and practices regarding health among tribes KAP Tribe χ p-value Total Akha Lahu Hmong Yao Karen Lisu n % n % n% n % n% n% n % Total 377 100.0 60 15.9 76 20.2 46 12.2 70 18.6 73 19.4 52 13.8 Knowledge Low 61 16.2 15 24.6 13 21.3 10 16.4 8 13.1 5 8.2 10 16.4 15.07 0.129 Moderate 167 44.3 24 14.4 33 19.8 21 12.6 38 22.8 31 18.6 20 12.0 High 149 39.5 21 14.1 30 20.1 15 10.1 24 16.1 37 24.8 22 14.8 Attitude Low 53 14.1 12 22.6 5 9.4 14 26.4 14 26.4 4 7.5 5 9.4 38.04 < 0.001* Moderate 250 66.3 44 17.6 55 22.0 25 10.0 42 16.8 44 17.6 40 16.0 High 74 19.6 4 5.4 16 21.6 7 9.5 14 18.9 25 33.8 8 10.8 Practice Low 47 12.5 3 6.4 8 17.0 10 21.3 9 19.1 8 17.0 9 19.1 10.51 0.397 Moderate 267 70.8 44 16.5 56 21.0 27 10.1 49 18.4 54 20.2 37 13.9 High 63 16.7 13 20.6 12 19.0 9 14.3 12 19.0 11 17.5 6 9.5 *Significance level at α=0.05 Table 7 Physical examination and medical history Item Total Male Female χ p-value n% n% n% BMI Underweight 116 14.6 62 53.4 54 46.6 3.98 0.137 Normal 320 40.4 168 52.5 152 47.5 Overweight 357 45.0 163 45.7 194 54.3 Disabled No 739 93.2 362 49.0 377 51.0 1.42 0.232 Yes 54 6.8 31 57.4 23 42.6 Heart disease No 724 96.1 337 46.5 387 53.5 0.37 0.538 Yes 29 3.9 16 55.2 13 44.8 History of TB diagnosis No 757 95.5 369 48.7 388 51.3 4.41 0.036* Yes 36 4.5 24 66.7 12 33.3 Sleeping problem No 674 85.0 356 52.8 318 47.2 19.09 < 0.001* Yes 119 15.0 37 31.1 82 68.9 Eye Normal 663 83.6 328 49.5 335 50.5 0.99 0.804 Cataract 77 9.7 36 46.8 41 53.2 Pterygium 50 6.3 27 54.0 23 46.0 History of glaucoma 3 0.4 2 66.7 1 33.3 Apidechkul BMC Public Health (2018) 18:694 Page 10 of 17 Table 7 Physical examination and medical history (Continued) Item Total Male Female χ p-value n% n% n% Tooth problem No 450 56.7 234 52.0 216 48.0 2.48 0.115 Yes 343 43.3 159 46.4 184 53.6 Headache No 557 72.1 302 54.2 275 49.4 6.55 0.010* Yes 216 27.9 91 42.1 125 57.9 Dizziness No 556 70.1 294 52.9 262 47.1 8.19 0.004* Yes 237 29.9 99 41.8 138 58.2 Peptic ulcer No 527 66.5 278 52.8 249 47.2 6.40 0.011* Yes 266 33.5 115 43.2 151 56.8 Anorexia No 707 89.2 371 52.5 336 47.5 22.18 < 0.001* Yes 86 10.8 22 25.6 64 74.4 History of injury No 713 89.9 349 48.9 364 51.1 1.05 0.305 Yes 80 10.1 44 55.0 36 45.0 History of hospital admission No 310 39.1 143 46.1 167 53.9 2.39 0.122 Yes 483 60.9 250 51.8 233 48.2 Parental history of DM No 515 64.9 262 50.9 253 49.1 1.01 0.313 Yes 278 35.1 131 47.1 147 52.9 Parental history of HT No 375 47.3 190 50.7 185 49.3 0.34 0.554 Yes 418 52.7 203 48.6 215 51.4 *Significance level at α=0.05 smoked. The participants had a high frequency of con- population, 21.9% did not know that they had HT. In sumption of dietary salt (64.4%), greasy food (62.5%), taking a closer look into tribal differences, more than sweet food (50.7%) and uncooked food (44.9%). Five fac- half of the Yao and Lisu participants had HT. This tors were found to be significantly associated with phenomenon could be attributed to the differences in T2DM: tribe, exercise, BMI, parental history of T2DM, culture and lifestyle among the hill tribe people, who and triglycerides. Another four factors were found to be consume alcohol and foods that are highly sweetened significantly associated with HT: sex, dietary salt intake, and salty and do not exercise regularly. BMI, and parental history of HT. In our study, the comorbidity rate of T2DM and HT is The results of our study revealed very interesting in- higher than that in an Indian sample in a study of Jaya formation on the prevalence of T2DM among the hill et al. [25]. However, the T2DM prevalence of our study tribe elderly populations in Thailand at 16.8%, which is sample is similar to that of a sample from a study con- 1.75 times higher than that of the Thai population [11]. ducted by Mohamed et al. [26] among the ethnic groups We also found significant differences in prevalence in northern Sudan, with a T2DM prevalence of 18.7%. among the various tribes. Meanwhile, the prevalence of Dhiraj et al. [27] reported that in different tribes of the HT was 45.5%, which is almost 1.6 times greater than population, there were different burdens of T2DM in that of the general Thai elderly population [13]. Among the sub-Himalayan region of India. This information the participants with HT in the hill tribe elderly supports the finding that the hill tribe people in Apidechkul BMC Public Health (2018) 18:694 Page 11 of 17 Table 8 Classification of participants’ characteristics by biomarkers 2 2 2 Factors Uric acid χ p-value Cholesterol χ p-value Triglyceride χ p-value Normal n (%) High n (%) Normal n (%) High n (%) Normal n (%) High n (%) Sex Male 246 (64.4) 136 (35.6) 38.63 <0.001* 286 (74.9) 96 (25.1) 14.28 <0.001* 309 (80.9) 73 (19.1) 2.44 0.118 Female 329 (83.9) 63 (16.1) 244 (67.4) 148 (32.6) 299 (76.3) 93 (23.7) Age (years) 60-69 311 (77.9) 88 (22.1) 6.04 0.049* 261 (65.4) 138 (34.6) 4.45 0.108* 303 (75.9) 96 (24.1) 6.58 0.037* 70-79 197 (71.1) 80 (28.9) 195 (78.9) 82 (21.1) 219 (79.1) 58 (20.9) ≥ 80 67 (68.4) 31 (31.6) 74 (78.7) 24 (21.3) 86 (87.8) 12 (12.2) Tribe Akha 101 (78.3) 28 (21.7) 20.19 0.018* 95 (73.6) 34 (26.4) 17.05 0.004* 99 (76.7) 30 (23.3) 8.86 0.114 Lahu 113 (85.0) 20 (15.0) 100 (75.2) 33 (24.8) 96 (72.2) 37 (27.8) Hmong 81 (64.3) 45 (35.7) 93 (73.8) 33 (26.2) 100 (79.4) 26 (20.6) Yao 96 (74.4) 33 (25.6) 82 (90.0) 47 (9.1) 98 (75.9) 31 (24.1) Karen 99 (76.7) 30 (23.3) 72 (55.8) 57 (44.2) 111 (86.0) 18 (14.0) Lisu 85 (66.4) 43 (33.6) 88 (68.8) 40 (31.2) 104 (81.3) 24 (18.7) *Significance level at α=0.05 Thailand originate from Tibet [14, 16], which is close to of T2DM, which is consistent with the findings of our those living in the sub-Himalayan region of India. study. Triglyceride levels are another factor related to Therefore, the T2DM and HT prevalence among the 6 the development of T2DM. A retrospective longitudinal hill tribes in Thailand are possibly different. large-scale study conducted between the year 2000 and A study using a mass database in Korea reported that 2012 found that every 10 mg/dL increase in triglyceride regular and frequent exercise led to reduced T2DM levels significantly increased the risk of T2DM by 4.0% mortality and morbidly rates, particularly in the elderly in the United States [37]. In addition, Ming et al. [38] re- population [28]. A study in Saudi Arabia also reported ported that an increase in triglycerides was a risk factor that sufficient physical exercise was a protective factor for type 2 diabetes among those living in rural China. against T2DM development [29]. This result is similar These studies present findings similar to those of this to the finding of our study that regular exercise and study, such that higher triglyceride levels are a risk fac- highly active physical work serve as protective factors tor for T2DM. Different tribes or races also have signifi- against T2DM among the hill tribe elderly populations cant associations with T2DM. The studies of Vitor [39] in Thailand. Regarding BMI, Kulaya et al. [30]reported and Diego et al. [40], which were conducted in the that increasing BMI was identified as a major risk fac- United States using different study designs, revealed that tor for T2DM in the Thai population. In a study of differences in the races of parents had an impact on the Asian Americans in the United States, a BMI< 23 or development of HT in their children. However, in our overweight was detected as a risk factor for T2DM de- study, there was no significant difference in HT preva- velopment [31]. Moreover, a case-control study aimed lence among the tribes. at assessing the association between BMI and T2DM in Jugal et al. [41] reported that there were several factors the Mid-Atlantic region found a heavy association be- associated with HT among those living in rural Delhi, tween increasing BMI and T2DM, after controlling for India, such as older age, alcohol use, education and chol- all confounding factors [32]. However, in a study esterol levels. However, sex was not found to be associ- among Afro-Trinidadians in the United States in 2016, ated with HT. On the other hand, Saswata et al. [42] no significant difference in BMI was found between reported that females had a greater chance of developing those who had T2DM and those who did not [33]. In HT than males in a study conducted in western India. our study, it was found that increasing BMI or over- Daily food consumption is one of the predictors for HT. weight was a risk factor for T2DM in the hill tribe Daily consumption of salty foods is one of the risk fac- elderly populations. tors of HT. This finding is supported by several studies Many studies [34–36] have reported that having a par- [43–45] that show that dietary salt intake was highly as- ental or family history of diabetes or first-degree rela- sociated with HT development in developing and devel- tives with diabetes was associated with the development oped countries and in urban and rural areas. In this Apidechkul BMC Public Health (2018) 18:694 Page 12 of 17 Table 9 Factors associated with T2DM in univariate and multivariate analyses (n = 775)** Factors T2DM OR 95%CI p-value OR 95%CI p-value adj Yes No n% n% Sex Mal 66 17.3 316 82.7 1.00 Female 64 16.3 329 83.7 0.93 1.02 -2.02 0.712 Tribe Akha 11 8.6 117 91.4 1.00 1.00 Lahu 26 19.5 107 80.5 2.58 1.37-4.85 0.013* 2.89 1.32-6.33 0.008* Hmong 11 8.1 124 91.9 0.94 0.45-1.96 0.896 0.91 0.35-2.31 0.845 Yao 26 21.5 95 78.5 2.91 1.54-5.48 0.006* 3.47 1.58-7.62 0.002* Karen 34 26.4 95 73.6 3.80 2.06-7.03 < 0.001* 5.03 2.35-10.78 < 0.001* Lisu 22 17.1 107 82.9 2.18 1.14-4.13 0.046* 2.73 1.22-6.07 0.014* Age (year) 60-69 75 18.8 324 81.2 1.00 70-79 39 14.3 234 85.7 0.72 0.50-1.02 0.127 ≥ 80 16 15.5 87 84.5 0.79 0.48-1.30 0.444 Smoking No 78 16.4 398 83.6 1.00 Ever in the past 34 23.1 113 76.9 1.53 1.04-2.24 0.064* Yes 18 11.8 134 88.2 0.68 0.43-1.08 0.177 Alcohol use No 79 15.0 447 85.0 1.00 Ever in the past 26 19.3 109 80.7 1.35 0.89-2.03 0.230 Yes 25 21.9 89 78.1 1.58 1.04 -2.42 0.072* Salty food No 151 53.5 131 46.5 1.00 Yes 266 52.1 245 47.9 0.94 0.70-1.26 0.687 Greasy food No 155 52.2 142 47.8 1.00 Yes 258 52.0 238 48.0 0.99 0.74-1.32 0.962 Sweet food No 202 51.7 189 48.3 1.00 Yes 184 45.8 218 54.2 0.78 0.59-1.04 0.097 Exercise No 21 26.6 58 73.4 1.00 1.00 Highly active physical work 45 16.7 225 83.3 0.55 0.33- 0.90 0.050* 0.48 0.25-0.91 0.024* Yes 64 15.0 362 85.0 0.48 0.30- 0.78 0.013* 0.45 0.24-0.83 0.011* BMI Normal 39 12.6 271 87.4 1.00 1.00 Underweight 13 11.4 101 88.6 0.89 0.51-1.56 0.743 0.90 0.45-1.80 0.773 Overweight 78 22.2 273 77.8 1.98 1.39- 2.82 0.001* 2.08 1.32-3.27 0.001* Parental history of DM No 217 42.1 298 57.9 1.00 1.00 Yes 149 53.6 129 46.4 1.58 1.18-2.12 0.002* 1.55 1.17-2.10 0.001* Apidechkul BMC Public Health (2018) 18:694 Page 13 of 17 Table 9 Factors associated with T2DM in univariate and multivariate analyses (n = 775)** (Continued) Factors T2DM OR 95%CI p-value OR 95%CI p-value adj Yes No n% n% Hypertension No 70 12.9 282 80.1 1.00 Yes 60 14.2 363 85.8 1.50 1.02- 2.19 0.035* Headache No 95 16.9 467 83.1 1.00 Yes 35 16.4 178 83.6 0.96 0.67-1.38 0.875 Dizziness No 86 15.9 456 84.1 1.00 Yes 44 18.9 189 81.1 1.23 0.88-1.72 0.303 Cholesterol Normal 90 17.3 430 82.7 1.00 High 38 16.1 198 83.9 0.91 0.64-1.29 0.682 Triglyceride Normal 88 14.9 504 85.1 1.00 1.00 High 40 24.4 124 75.6 1.84 1.29-2.63 0.004* 1.73 1.10-2.73 0.017* *Significance level at α=0.05 **18 participants could not provide blood specimens study, we also found that dietary salt intake among the may have occurred because they clearly did not under- hill tribe elderly populations was a significant risk factor stand the importance of laboratory interpretations. for HT development. Another factor related to HT is Moreover, most hill tribe elderly adults are not edu- BMI. Alicja et al. [46] reported that both men and cated. This finding coincides with those of studies by women had an increased risk of HT with increasing Apidechkul et al. [53]and Apidechul [54], who re- BMI, particularly among the elderly populations. A rural ported that a high proportion of the Akha elderly popu- Chinese cohort study in 2016 [47] and a study in lation and the Lahu people were in the illiterate group. Bangladesh in 2017 [48] confirmed that the increase in This finding could explain participants’ limited under- BMI had a significant association with HT development. standing of the research information and lack of These findings coincide with those of our study, which cooperation with the procedure. revealed that an increase in BMI was associated with a The researchers could not draw blood from a few partic- greater odds of HT development among the hill tribe ipants (1.26%) because of their individual peripheral vein elderly populations in Thailand. characteristics. However, nobody refused to provide infor- The study of Ghada et al. [49] in Egypt showed a mation and a specimen. Because this lack of data would strong association between a family history of HT and affect the predictive statistical model (logistic regressions), the development of HT in one’s offspring. A family his- these participants were excluded from the analysis to en- tory has been detected as a risk factor for HT among sure the accuracy of the results. Furthermore, some partic- young adults and the elderly population in several ipants had been diagnosed as T2DM and HT before countries [50–52]. starting the study, which could possibly impact the find- Some limitations have been identified in this study, ings of the study, particularly their knowledge, attitudes such as misunderstanding the NPO techniques before and practices, which are common limitations of the drawing blood specimens, language, and the inability to cross-sectional study design. Concerning this point, know- draw blood specimens in some people. Since some tar- ledge of and attitudes toward DM and HT were not in- geted hill tribe villages are located far away from the cluded in the prediction model. Moreover, if we look city, traveling to the study setting very early in the day closely, only attitude is significantly different among the to collect blood specimens was sometimes not practical. tribes. Additionally, the number of Lahu (excess of 3par- Other limitations included unclear information on the ticipants) and Hmong (excess of 10 participants) partici- research procedure and not drinking and eating food for pants exceeded the minimum requirement for the sample at least 8 hours before having blood drawn. Sometimes size due to miscommunication between the researcher there was no cooperation from the participants, which and community headman. However, these excess data did Apidechkul BMC Public Health (2018) 18:694 Page 14 of 17 Table 10 Factors associated with HT in univariate and multivariate analyses Factors HT OR 95%CI p-value OR 95%CI p-value Adj Yes No n% n% Sex Male 164 41.7 229 58.3 1.00 1.00 Female 197 49.3 203 50.7 1.35 1.02-1.79 0.034* 1.29 1.01-1.68 0.031* Tribe Akha 61 46.9 69 53.1 1.00 Lahu 61 45.9 72 54.1 0.95 0.59-1.55 0.863 Hmong 42 30.0 98 70.0 0.48 0.29-0.79 0.004* Yao 74 56.9 56 43.1 1.49 0.91-2.43 0.107 Karen 52 40.0 78 60.0 0.75 0.46-1.23 0.261 Lisu 71 54.6 59 45.4 1.36 0.83-2.21 0.215 Age (years) 60-69 173 42.2 237 57.8 1.00 70-79 134 48.0 145 52.0 1.26 0.93-1.71 0.131 ≥80 54 51.9 50 48.1 1.48 0.96-2.27 0.075 Smoking No 233 47.9 253 52.1 1.00 Ever in the past 65 43.0 86 57.0 0.82 0.56-1.18 0.293 Yes 63 40.4 93 59.6 0.73 0.51-1.06 0.100 Alcohol use No 249 46.3 289 53.7 1.00 Ever in the past 64 46.0 75 54.0 0.99 0.68-1.44 0.960 Yes 48 41.4 68 58.6 0.81 0.54-1.23 0.337 Salty food No 138 48.9 144 51.1 1.00 1.00 Yes 307 60.1 204 39.9 1.57 1.17-2.01 0.002* 1.48 1.14-2.00 0.001* Greasy food No 136 45.8 161 54.2 1.00 Yes 241 48.6 255 51.4 1.11 0.83-1.49 0.582 Sweet food No 202 51.7 189 48.3 1.00 Yes 197 49.0 205 51.0 0.89 0.68-1.18 0.454 Regular Exercise Yes 36 45.0 44 55.0 1.00 Highly active physical work 113 40.5 166 59.5 0.83 0.50-1.37 0.472 No 212 48.8 222 51.2 1.16 0.72-1.88 0.527 BMI Normal 112 35.0 208 65.0 1.00 1.00 Underweight 42 36.2 74 63.8 1.05 0.67-1.64 0.816 2.56 0.70 – 1.70 0.696 Overweight 207 58.0 150 42.0 2.56 1.87- 3.49 < 0.001* 1.37 1.01 – 1.90 < 0.001* Parental history of HT No 155 41.3 220 58.7 1.00 1.00 Yes 302 72.2 116 27.8 3.69 2.74-4.97 < 0.001* 3.38 2.81-4.48 < 0.001* Apidechkul BMC Public Health (2018) 18:694 Page 15 of 17 Table 10 Factors associated with HT in univariate and multivariate analyses (Continued) Factors HT OR 95%CI p-value OR 95%CI p-value Adj Yes No n% n% Diabetes mellitus No 282 43.7 363 56.3 1.00 Yes 70 53.8 60 46.2 1.50 1.02- 2.19 0.035* Headache No 249 43.2 328 56.8 1.00 Yes 112 51.9 104 48.1 1.41 1.03-1.94 0.029* Dizziness No 238 42.8 318 57.2 1.00 Yes 123 51.9 114 48.1 1.44 1.06-1.95 0.019* Cholesterol Normal 238 44.9 292 55.1 1.00 High 115 47.1 129 52.9 1.09 0.80-1.48 0.564 Triglyceride Normal 262 43.1 346 56.9 1.00 High 91 54.8 75 45.2 1.60 1.13- 2.26 0.007* *Significance level at α=0.05 not impact the results of study but rather supported the Acknowledgements The author would like to thank all the participants for kindly providing all power of the tests. essential information regarding the research procedures. The author is also Conducting research with the hill tribe people, grateful to all research assistants from the Center of Excellence for the Hill particularly among the elderly population, required tribe Health Research for their help in data collection. The author would like to thank The National Research Council of Thailand and Mae Fah Luang researchers to be clearly knowledgeable about the condition University, Thailand in support the grant. before reaching them. Additionally, having research assis- tants who were fluent in both Thai and the local hill tribe Funding languages was an advantage for obtaining information. This research was supported by the National Research Council of Thailand and Mae Fah Lung University, Thailand (Grant Number 77-2015). Conclusions The hill tribe elderly populations in Thailand are living Availability of data and materials The raw data supporting these findings can be found in the Additional file 1. with a high burden of T2DM and HT. T2DM and HT screening programs in these populations should be imple- mented regularly to detect early-stage and new cases. Authors’ contributions TA sought funding, designed the study protocols and procedures, collected There is an urgent need to develop proper health behavior data, analyzed and interpreted data, drafted, revised, and approved the final change models to reduce BMI and the consumption of version of the manuscript. dietary salt and greasy foods among the elderly popula- tions. Moreover, a program to encourage physical exercise Ethics approval and consent to participate is also necessary. Otherwise, Thailand must budget large Consent to participate, all study instruments and procedures were approved by amounts of money to provide care and treatment for these the Ethics Committee for Human Research, Mae Fah Laung University, Chiang Rai, Thailand (No. REH-58087). All participants received an oral and written populations in the near future. explanation and provided their consent before a voluntary agreement was witnessed and documented by signature or fingerprint. Additional file Competing interests Additional file 1: Hill tirb Elderly Data. (XLSX 545 kb) The author declares that he has no competing interests. Abbreviations BMI: Body mass index; DALYS: Disability adjusted life year; HT: Hypertension; Publisher’sNote ID: Identification; IOC: Item objective congruence; NPO: Nothing per oral; Springer Nature remains neutral with regard to jurisdictional claims in T2DM:: Type 2 diabetes mellitus; WHO: World Health Organization published maps and institutional affiliations. Apidechkul BMC Public Health (2018) 18:694 Page 16 of 17 Received: 1 February 2018 Accepted: 24 May 2018 24. American Association of Clinical Endocrinologists and American College of Endocrinology. Guidelines for management of dyslipdemia and prevention of cardiovascular disease. 2017. https://www.aace.com/files/ lipid-guidelines.pdf 25. Jaya PT, Thakur JS, Gursimer J, Sanjay J. Prevalence and determinants of References comorbidity diabetes and hypertension: evidence from non communicable 1. World Health Organization (WHO). Country and region data on diabetes disease risk factor STEPS survey, India. Diabetes & Metabolic Syndrome: and hypertension, 2017. 2017. http://www.who.int/diabetes/facts/world_ Clinical Research and Reviews. 2017;11(Sup1):S459–65. figures/en/index2.html 26. Mohamed AE, Abubakr BM, Hind EE, Kamal Y, Sufian KN, Waldie ME, et al. 2. World Health Organization (WHO). Fact sheet: Diabetes mellitus November Increasing prevalence of type 2 diabetes mellitus and impact of ethnicity in 2017. 2017. http://www.who.int/mediacentre/factsheets/fs312/en/ north Sudan. Diabetes Research and Clinical Practice. 2018;136:93–9. 3. Centers for Disease Control and Prevention (CDC). High blood pressure risk 27. Dhiraj K, Ashok KB, Dinesh K, Sunil KR. Prevalence of diabetes mellitus and factors 2017. 2017. https://www.cdc.gov/bloodpressure/risk_factors.htm its risk factors among permanently settled tribal individuals in tribal and 4. Sinna PU, Anil M, Elisabeth S, Finn BL, Reimar WT. Lifestyle factors associated urban areas in northern state of Sub-Himalayan Region of India. with type 2 diabetes and use of different glucose-lowering drugs: Cross- International of Chronic Diseases. 2014; Article ID 380597, 9 pages. https:// sectional study. PLOS ONE. 2014;9(11):e111849. https://doi.org/10.1371/ doi.org/10.1155/2014/380597. journal.pone.0111849 28. Woo YS, Taehee L, Da-Hye J, Hyeon CK. Diabetes, frequency of exercise, and 5. Funda DY, Zeynel AO. Treatment of type 2 diabetes mellitus in the elderly. mortality over 12 years: Analysis of the National Health Insurance Service- World J Diabetes. 2017;8(6):278–85. Health Screening (NHIS-HEALS) Database. J Korean Med Sci. 2018;33(8):e60. 6. Peng L, Jin-Gang Z. Blood pressure targets in the hypertensive elderly. Chin 29. Alzaheb RA, Altemani AH. The prevalence and determinants of poor Med J. 2017;130(16):1968–72. glycemic control among adults with type 2 diabetes mellitus in Saudi 7. Davies A, Janet OI, Adewale VA, Ngozi A, Emmanuel OA, Asa A, Gbolahan O. Arabia. Diabetes Metab Syndr Obes. 2018;11:15–21. Estimating the prevalence, hospitalization and mortality from type 2 30. Kulaya N, Natkamol C, Panuwat P, Jariya H. Screening high-risk populations diabetes mellitus in Nigeria: a systematic review and meta analysis. BMJ for hypertension and type 2 diabetes among Thais. WHO South-East Asia Open. 2017;7:e015424. https://doi.org/10.1136/bmjopen-2016-015424. Journal of Public Health. 2012;1(3):320–9. 8. Yashikin AP, Kravchenko J, Yashin AI, Sloan F. Mortality and macrovascular 31. Karen MK, Keith TC, Esme FT. Diabetes among Asian Americans with BMI risk in elderly with hypertension and diabetes: effect of intensive drug less than or equal to 23. Diabetes& Metabolic Syndrome: Clinical Research & therapy. American Journal of Hypertension. 2018;31(2):220–7. Reviews. 2017; https://doi.org/10.1016/j.dsx.2017.12.011 9. Sophia DF, Sunny DF. Economic burden of diabetes mellitus and its socio- 32. Michael LG, Neil W, Qian L, Veronica A, Jakob L, Mette H. The association of economic impact on household expenditure in an urban slum area. Int J body mass index with the risk of type 2 diabetes: a case-control study nested Res Med Sci. 2017 May;5(5):1808–13. in an electronic health records system in the United States. BMC Diabetology 10. World Health Organization (WHO). Diabetes country profiles, 2026: Thailand & Metabolic Syndrome. 2014;6:50. https://doi.org/10.1186/1758-5996-6-50 diabetes. 2016. http://www.who.int/diabetes/country-profiles/tha_en.pdf. 33. Shivananda N, Kheleel S, Shivang S, Vashista S, Rukaiya S, Safayah S, et al. 11. Chaicharn D, Alessandra F. Diabetes management in Thailand: a literature Investing the link between benign prostatic hypertrophy, BMI, and type 2 review of the burden, cost, and outcomes. BMC Global Health. 2013;9(11) diabetes mellitus. Diabetes& Metabolic Syndrome: Clinical Research Reviews. https://doi.org/10.1186/1744-8603-9-11. 2017;11(Siup2):S627–30. 12. World Health Organization (WHO). Raised blood pressure: Situation and 34. Deepark KD, Arun KC, Mamta S, Swayamsidha M, Binoy KM, Anoj KB. Effect trend. 2017. http://www.who.int/gho/ncd/risk_factors/blood_pressure_ of parental history of diabetes on markers of inflammation insulin resistance prevalence_text/en/. and atherosclerosis in first degree relatives of patients with type 2 diabetes 13. World Health Organization (WHO)-Thailand. Technical consultation to mellitus. Diabetes & Metabolic Syndrome: Clinical Research & Reviews. 2017; strengthen hypertension care in Thailand, 2017. 2017. http://www.searo. https://doi.org/10.1016/j.dsx.2017.12.004 who.int/thailand/news/htn-meeting-report-draft24nov.pdf?ua=1 35. Kara MG, Shannon KC, Duck-chul L, Xuemei SS, Steven PH, Steven NB. 14. Princess Maha Chakri Siridhorn Anthropology center. Hill tribe. 2018. Association of cardiorespiratory fitness and parental history of diabetes with risk http://www.sac.or.th/main/index.php of type 2 diabetes. Diabetes research and Clinical Practice. 2012;95(3):425–31. 15. The hill tribe welfare and development center, Chiang Rai province. Hill 36. Chaochen W, Hiroshi Y, Koji T, Hideaki T, Keiko W, Yuanying L, et al. tribe population. The hill tribe welfare and development center. Chiang Rai: Association between parental history of diabetes and the incidence of type Ministry of Interior; 2017. p. 22–31. 2 diabetes mellitus differs according to the sex of the parent and offspring’s 16. Tawatchai A. A 20-year retrospective cohort study of TB infection among body weight: A finding from a Japanese worksite based cohort study. the hill tribe HIV/AIDS populations, Thailand. BMC Infectious Disease. 2016; Preventive Medicine. 2015;81:49–53. 16:72. https://doi.org/10.1186/s12879-016-1407-4. 37. Aman B, Eytan C, Elad G, Pearl L, Moshe G, Ilan K. Triglyceride levels and risk 17. Tawatchai A, Onn L, Sirinan S. Inequity in accessing health care service in of type 2 diabetes mellitus: A longitudinal large study. J Investig Med. 2016; Thailand in 2015: A case study of the hill tribe people in Mae Fah Luang 64(2):383–7. district, Chiang Rai, Thailand. J Health Res. 2016;30(1):67–71. 38. Ming Z, Bingyuan W, Yu L, Xizhuo S, Xinping L, Chongjian W, et al. 18. Sudkhed D, Tawatchai A, Rachanee S, Amornrat A, Onnalin S, Bukhari P, Cumulative increased risk of incident type 2 diabetes mellitus with et al. Patterns and perception of alcohol drinking among the Lahu people, increasing triglyceride glucose index in normal-weight people: the rural northern Thailand. J Health Res. 2018;32(3):67–71. Chinese cohort. BMC Cardiovascular Diabetology. 2017;16(30): https://doi. 19. Incidence and risk factors for type 2 diabetes mellitus in transitional org/10.1186/s12933-017-0514-x Thailand: results from the Thai cohort study. BMJ Open. 2016;6:e014102. https://doi.org/10.1136/bmjopen-2016-014102 39. Victor G. Pregravid hypertension may have different secondary sex ratio 20. World Health Organization (WHO). Definition and diagnosis of diabetes effects in different races in the United States. International Journal of mellitus and intermediate hyperglycemia. 2017. http://www.who.int/ Cardiology. 2017;249:333–5. diabetes/publications/Definition%20and%20diagnosis%20of%20diabetes_ 40. Diego GD, Arthur EM, Raphael MR, Romulo AF, Bruna TC, Mariana RP, et new.pdf al. Association between hypertension in adolescents and the health risk factors of their parents: an epidemiological family study. Journal of 21. The hill tribe welfare and development center, Chiang Rai province. Hill American Society of Hypertension. 2018; https://doi.org/10.1016/j.jash. tribe population. In: The hill tribe welfare and development center. Chiang 2017.12.011 Rai: Ministry of Interior; 2015. p. 16–27. 41. Jugal K, Neeru G, Charu K, Neeta K. Prevalence of hypertension and 22. American College of Cardiolog. 2017 Guideline for prevention, determination of its risk factors in rural Delhi. International Journal of detection, evaluation, and management of high blood pressure in Hypertension. 2016; https://doi.org/10.1155/2016/7962595. adult. file:///Users/macbookpro13/Downloads/Guidelines_Made_Simple_ 2017_HBP.pdf 42. Saswata G, Simantini M, Anamitra B. Sex difference in the risk profile of 23. World Health Organization (WHO). BMI classification in Asian population. hypertension: A cross-sectional study. BMJ open. 2016:e010085. https://doi. 2015. http://apps.who.int/bmi/index.jsp?introPage=intro_3.html org/10.1136/bmjopen-2015-010085. Apidechkul BMC Public Health (2018) 18:694 Page 17 of 17 43. Sung KH. Dietary salt intake and hypertension. Electrolyte Blood Press. 2014; 12(1):7–18. 44. William BF, David GE, Claudine TJ, William SW. Dietary sodium and health: more than just blood pressure. J Am Coll Cardiol. 2015;65(10):1042–50. 45. Mandapaka RT, Rachabathuni S. Prevalence of hypertension and its relationship between dietary salt intake in urban population. J Community Med Health. 6:426. https://doi.org/10.4172/2161-0711.1000426. 46. Alicja S, Anna L. BMI, hypertension and low bone mineral density in adult men and women. HOMO-Journal of Comparatie Human Biology. 2012;63(4):282–91. 47. Ming Z, Yang Z, Haohang S, Xinping L, Chongjian W, Linlin L. Effect of dynamic change in body mass index on the risk of hypertension: results from the rural Chinese chohort study. International Journal of Cardiology. 2017;238:117–22. 48. Mohammed MR, Shamima A, Jenny J, Mohammed SR, Papia S. Trend, projection, and appropriate body mass index cut-off point for diabetes and hypertension in Bangladesh. Diabetes Research and Clinical Practice. 2017; 126:43–53. 49. Ghada Y, Ibrahim ET, Dina O, Ahmed S, Essam B, Zeinab A, et al. Familial history of hypertension as a predictor of increased arterial stiffness in normotensive offspring. The Egypt Heart Journal. 2017;69(1):37–44. 50. Menghetti E, Strisciuglio P, Spagnolo A, Carletti M, Paciotti G, Muzzi G, et al. Hypertension and obesity in Italian school children: The role of diet, life style and family history. Nutrition, metabolism and Cardiovascular Diseases. 2015;25(6):602–7. 51. Dorairaj P, Panniyammakal J, Shreeparna G, Roopa S, Vamadevan SA, Dimple K, et al. Prevalence and incidence of hypertension: results from a representative cohort of over 16,000 adults in three cities of South Asia. Indian Heart Journal. 2017;69(4):434–41. 52. Mercedes RC, Guadalupe XA, Shrikant IB. Association of cardiovascular risk factors between Hispanic/Latino parents and youth: the Hispanic community health study/study of Latino youth. Annals of Epidemiology. 2017;27(4):260–8. 53. Tawatchai A, Pilasinee W, Siriyaporn S, Thapakorn R. Health situation of Akha hill tribe in Chiang Rai province, Thailand. Journal of Public Health and Development. 2016;14(1):77–97. 54. Tawatchai A. Prevalence of thalassemia carriers among the Lahu hill tribe population, Chiang Rai, Thailand. Asian Biomedicine. 2015;9(4):527–33.

Journal

BMC Public HealthSpringer Journals

Published: Jun 5, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off