- Email: [email protected]
The risk factors for gestational diabetes mellitus: A retrospective study
Midwifery 42 (2016) 16–20
Contents lists available at ScienceDirect
Midwifery journal homepage: www.elsevier.com/midw
The risk factors for gestational diabetes mellitus: A retrospective study Pei-Chao Lin, PhD, RN (Assistant professor)a, Chich-Hsiu Hung, PhD, RN (Professor)a,n, Te-Fu Chan, PhD, MD (Associate professor)b,c, Kuan-Chia Lin, PhD (Professor)d, Yu-Yun Hsu, PhD, RN (Associate professor)e, Ya-Ling Tzeng, PhD, RN (Professor & Director)f a
School of Nursing, Kaohsiung Medical University, No. 100, Shih-Chuan 1st Rd., Kaohsiung City 80708, Taiwan Department of Obstetrics and Gynecology, School of Medicine, Kaohsiung Medical University, No. 100, Shih-Chuan 1st Rd., Kaohsiung City 80708, Taiwan c Department of Obstetrics and Gynecology, Kaohsiung Medical University Chung-Ho Memorial Hospital, No. 100, Tz-You 1st Rd., Kaohsiung City 80756, Taiwan d Institute of Hospital and Health Care Administration, National Yang-Ming University, No. 155, Linong Street Sec. 2, Taipei City 11221, Taiwan e Department of Nursing, College of Medicine, National Cheng Kung University, 1 University Road, Tainan City, 70101, Taiwan f School of Nursing and Graduate Institute of Nursing, China Medical University, No. 91, Hsueh-Shih Rd., Taichung City 40402, Taiwan b
art ic l e i nf o
a b s t r a c t
Article history: Received 25 January 2016 Received in revised form 15 July 2016 Accepted 21 September 2016
Objective: To investigate the risk factors for developing GDM among Taiwanese pregnant women. Design: A retrospective cohort and case-control study. Setting: At a medical centre in Southern Taiwan. Participants: The hospitalised pregnant women who were diagnosed with either GDM or normal glucose tolerance (NGT) between 1997 and 2011. The glucose tolerance test results were interpreted according to criteria established by the National Diabetes Data Group for GDM. Participants were divided into either a GDM group (case group) or a normal glucose tolerance (NGT) group (control group) in order to determine the risk factors for GDM. Measurements: With a retrospective chart review, data regarding demographics, a family history of diabetes, history of gestation, and physiological index for pre- and postpregnancy periods were collected. χ2 tests and independent t tests were used to examine the correlations between demographic characteristics and GDM. Stepwise multivariate logistic regression was used to determine the factors associated with GDM. Findings: The results of the comparison between the GDM group (n ¼ 106) and the NGT group (n ¼ 406) showed that the risk factors for GDM were maternal age, education, a family history of diabetes, and prepregnancy body mass index (BMI). Key conclusion and implication for practice: Older age, lower levels of education, a family history of diabetes, and higher prepregnancy BMI were significant risk factors for GDM. In addition to performing risk factor assessment, health care providers should proactively promote the importance of GDM screening to pregnant women at their first antenatal visit. & 2016 Elsevier Ltd. All rights reserved.
Keywords: Age Body mass index Education status Gestational diabetes mellitus Taiwan
Introduction Gestational diabetes mellitus (GDM) is a major global public health concern, with prevalence increasing yearly. Women with diabetes have high risk to give birth to giant babies and to suffer from mortality (Ryan, 2013). GDM can result in serious maternal, foetal, and neonatal health implications, including preeclampsia,
n
Corresponding author. E-mail addresses: [email protected] (P.-C. Lin), [email protected] (C.-H. Hung), [email protected] (T.-F. Chan), [email protected] (K.-C. Lin), [email protected] (Y.-Y. Hsu), [email protected] (Ya-Ling Tzeng). http://dx.doi.org/10.1016/j.midw.2016.09.008 0266-6138/& 2016 Elsevier Ltd. All rights reserved.
shoulder dystocia, macrosomia, neonatal hypoglycaemia, etc (Mitanchez, 2010). The reported prevalence of GDM worldwide ranges from 2% to 6%; the prevalence of GDM in India, the Middle East, and Sardinia ranges from 10% to 22% (Galtier, 2010). Asian women appear to be at a higher risk of developing GDM than non-Hispanic white women (Savitz et al., 2008). The prevalence of GDM in China was 4.3% (Yang et al., 2009), and a recent study reported a GDM prevalence of up to 7.4% in Taiwan (Lin et al., 2009). NonHispanic Asian/Pacific Islander women were 2.26-fold more likely to develop GDM compared with non-Hispanic white women (Hunsberger et al., 2010). GDM not only adversely affects maternal and child health but also increases the resulting medical costs (Chen et al., 2009). Several studies have reported that women with
P.-C. Lin et al. / Midwifery 42 (2016) 16–20
a history of GDM are at an increased risk of type 2 diabetes (Golden et al., 2009) and cardiovascular diseases (King et al., 2009). Therefore, health care providers should understand the risk factors for GDM and provide pregnant women with preventive care and early intervention. Maternal age is widely considered one of the risk factors associated with GDM (Dode and Santos, 2009). The risk of developing GDM increases with maternal age and an increasing prepregnancy body mass index (BMI) (Dode and Santos, 2009). The results of a study demonstrated that women with lower levels of education had a higher risk of developing GDM than that of women with higher levels of education (van der Ploeg et al., 2011). However, another study found no association between GDM and education in Chinese pregnant women (Yang et al., 2009). Numerous studies have demonstrated that women with a family history of diabetes had a higher risk of developing GDM than that of those without (van Leeuwen et al., 2010; Rhee et al., 2010). Dode and Santos (2009) reviewed relevant literature and discovered that, of the 23 articles addressing the associations between GDM and parity, only five present positive associations. Most studies have indicated that a short stature was positively associated with GDM (Dode and Santos, 2009); however, we cannot rule out the possibility of selection bias on these studies. Despite the high prevalence of GDM in Asia, few studies have focused on the risk factors for GDM, and information about factors associated with GDM among the Asian population is limited. Additional research is required to validate previous research on the factors associated with GDM. Therefore, the purpose of this study was to investigate the risk factors for developing GDM among Taiwanese pregnant women.
Methods Design This study used a retrospective chart review to collect data. The glucose tolerance test results were interpreted according to criteria established by the National Diabetes Data Group for GDM. A case-control study was conducted to determine the risk factors for GDM. Participants were divided into either a GDM group (case group) or a normal glucose tolerance (NGT) group (control group). Participants and procedure In our study, the inclusive criteria were women who were diagnosed with GDM, and gave birth and were discharged from the study hospital. The exclusive criteria involving women's medical records indicated incomplete glucose tolerance test, women with a diagnosis of type 1 diabetes or type 2 diabetes before pregnant. The minimum sample size for logistic regression was calculated according to a artificial milk (Hsieh et al., 1998) with the study results of Yang et al. (2009). If the statistical power (1 β) was 0.80, α was 0.05, and the number of participants in the control group was triple for the number of participants in the case group, the minimum sample size was 123 participants in the case group and 369 participants in the control group. This study was conducted at a medical centre in Southern Taiwan. Prior to data collection, the study was reviewed and approved by the institutional review board of the study hospital (KMUHIRB-2012-03-08(I)). Data were collected from the medical records of hospitalised women who delivered in and then were discharged from the study hospital between 1997 and 2011 with the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnostic codes of 648 (n ¼188) and 650 (n ¼678). A total of 866 medical records of patients met the
17
inclusion criteria, of which 122 could not be retrieved. After reviewing the available medical records of 744 pregnant women, we excluded 197 women who lacked data on the glucose tolerance test, antenatal visits, or child delivery. Furthermore, we excluded 10 women with a diagnosis of type 1 diabetes, 14 with type 2 diabetes, and 11 with impaired glucose tolerance. Eventually, 512 patients, 106 whose first diagnosis was GDM and 406 whose first diagnosis was NGT, were included in the analysis (Fig. 1). Data analysis Descriptive statistics were used to present the demographic characteristics of the participants. χ2 tests and independent t tests were used to examine the correlations between demographic characteristics and GDM. Stepwise multivariate logistic regression was used to analyse the risk factors for GDM. IBM SPSS Statistics Version 20 (Mandarin Chinese Edition) was used for data filing and analysis.
Findings The results of comparing the GDM group (n ¼106) with the NGT group (n¼ 406) showed significant differences (po .05) in mean age, education, employment status, a family history of diabetes, parity, body height, and prepregnancy BMI (Table 1). The results of the univariate logistic regression analysis indicated that the risk of developing GDM increased by 13% (95% confidence interval [CI]: 1.07–1.19, p o.001) for every year of maternal age, that women with an education level of junior college or below had a 3.28-fold higher risk (95% CI: 2.11–5.10, po .001) of GDM than that of women with an education level of university or above, and that unemployed women had a 1.93-fold higher risk (95% CI: 1.25–2.99, p¼ .003) of GDM than that of employed women. Additionally, women with a family history of diabetes exhibited a 7.16-fold higher risk (95% CI: 4.50–11.41, po .001) of GDM than that of women without a family history of diabetes; multipara had a 1.93-fold higher risk (95% CI: 1.24–3.02, p ¼ .004) of GDM than that of primipara. The risk of GDM was reduced by 5% (95% CI: 0.91–0.99, p ¼.008) for every 1-cm increase in body height; and the risk of GDM increased by 35% (95% CI: 1.26–1.46, po .001) for every 1 kg/m2 increase in prepregnancy BMI (Table 2). The results of the stepwise multivariate logistic regression analysis indicated that only four variables (age, education, a family history of diabetes, and prepregnancy BMI) were entered into the model. After adjusting for other variables, we determined that the risk of developing GDM increased by 10% (95% CI: 1.04–1.17, p¼ .002) for every year of maternal age; women with an education level of junior college or below had a 3.59-fold higher risk (95% CI: 2.07–6.21, p o.001) of GDM than that of women with an education level of university or above; women with a family history of diabetes showed a 6.79-fold higher risk (95% CI: 3.92–11.76, p o.001) of GDM than that of women without; and the risk of GDM increased by 29% (95% CI: 1.19–1.41, p o.001) for every 1 kg/m2 increase in prepregnancy BMI. To summarise, age, education, a family history of diabetes, and prepregnancy BMI were significant predictors of GDM (Table 3).
Discussion The results indicate that Taiwanese women's age, education, family history of diabetes, and prepregnancy BMI were risk factors for developing GDM. Evidence-based data reported suggested that older maternal age increased a woman's risk of developing GDM and this increase was most evident among women born in Asia
18
P.-C. Lin et al. / Midwifery 42 (2016) 16–20
A total of 866 hospitalized women who delivered in and were discharged from the study hospital between 1997 and 2011 with the ICD-9-CM diagnostic codes of 648 (n = 188) and 650 (n = 678)
343 women were excluded: 1. 122 medical records could not be retrieved 2. 10 medical records showed women with a diagnosis of type 1 DM before pregnant 3. 14 medical records indicated women with a diagnosis of Type II DM before pregnant 4. 197 medical records did not show women with any data for the glucose tolerance test, prenatal visits, or child delivery
523 women were undertaken glucose screening test 11 women were excluded due to impaired glucose tolerance 512 women included in the study
106 women with GDM
406 women with NGT Fig. 1. Flow chart of study participants.
Table 1 Demographic Characteristics of Women with and without GDM (n ¼ 512). Demographic characteristics
Age (year) Education Junior college or above Senior high or below Employment status Yes No Family history of diabetes Yes No Parity 0 1–3 Body height (cm) Prepregnancy BMI
Women with GDM mean7 SD n (%)
Women with NGT mean 7SD n (%)
T value or χ2 value
P-value
32.2 7 5.0
30.0 7 3.9
4.21 40.42
o .001 o .001
59(55.7)
342(84.2)
47(44.3)
64(15.8)
57(53.8) 49(46.2)
281(69.2) 125(30.8)
66(62.3) 40(37.7)
76(18.7) 330(81.3)
62(58.5) 44(41.5) 158.8 7 5.7 24.4 74.6
297(73.2) 109(28.8) 160.3 75.1 20.2 7 2.6
8.93
.003
79.52
o .001
8.62
.003
2.69 7.57
.007 o .001
and the Middle East (Carolan et al., 2012). The current study also demonstrated that older maternal age was associated with an increased risk of developing GDM. Therefore, health care providers should not only identify pregnant women with advanced maternal age as a high-risk group and provide them with related preventive health intervention but also proactively promote awareness of GDM and encourage older pregnant women to receive GDM
Table 2 Univariate logistic regression analysis for factors associated with GDM (n¼ 512). Demographic characteristics
OR
95% CI
P-value
Age Education, n (%) Junior college or below University or above Employment status Yes No Family history of diabetes Yes No Parity 2–3 1 Body height Prepregnancy BMI
1.13
(1.07–1.19)
o .001
3.28 1.00
(2.11–5.10)
o .001
(1.25–2.99)
.003
7.16 1.00
(4.50–11.41)
o .001
1.93 1.00 0.95 1.35
(1.24–3.02)
.004
(0.91–0.99) (1.26–1.46)
.008 o .001
1.00 1.93
screenings. Taiwanese women with an education level of junior college or below had a 3.59-fold higher risk of GDM than that of women with an education level of university or above. This finding is similar to those reported by previous studies, demonstrating that women with a low education level had an increased risk of GDM compared with women with a high education level (Hedderson and Ferrara, 2008; van der Ploeg et al., 2011). However, Khan et al. (2013) suggested that education level is not associated with GDM. Although sufficient evidence-based information about the association between GDM and education is scant, education appears to
P.-C. Lin et al. / Midwifery 42 (2016) 16–20
Table 3 Stepwise multivariate logistic regression analysis for factors associated with GDM (n¼ 512). Demographic characteristics
Adjusted OR
95% CI
P-value
Age Education Junior college or below University or above Family history of diabetes Yes No Prepregnancy BMI
1.10
(1.04–1.17)
.002
3.59 1.00
(2.07–6.21)
o.001
6.79 1.00 1.29
(3.92–11.76)
o.001
(1.19–1.41)
o.001
have a favourable impact on women's health conditions (Koch et al., 2013). A low education level may adversely affect a person's nutrition knowledge (Parmenter et al., 2009). For women with a low level of education who are at risk of developing GDM, health care providers should provide nutrition knowledge during pregnancy, thereby helping pregnant women maintain proper nutrition and ensure maternal and child health. The results of the univariate logistic regression analysis suggest that Taiwanese unemployed women were at high risk of GDM. This finding is consistent with that of Khan et al. (2013). Few studies have addressed the association between employment status and GDM. However, a previous study discovered that women with a lower socioeconomic status had an increased risk of GDM (Anna et al., 2008). In the current study, after adjustment for other variables including age, education, family history, and prepregnancy BMI, there were no significant differences in the association of employment status with GDM. Nevertheless, based on previous study findings, we suggest that health care providers pay particular attention to health risk assessment and management during pregnancy among women with low socioeconomic status. Among all the risk factors identified in our study, a family history of diabetes was the strongest predictor of GDM. Our finding regarding the risk ratio of a family history of diabetes for GDM is congruent with those reported in previous Asian studies (Rhee et al., 2010; van Leeuwen et al., 2010; Yang et al., 2011). After adjustment for other risk factors, the current study derived a higher risk ratio than those of previous Asian studies. A family history of diabetes was the major risk factor for GDM in Taiwan (Kuti et al., 2011). Additionally, Ogonowski et al. (2014) suggested that women with a matrilineal family history of diabetes were at a higher risk of GDM than women with a patrilineal family history of diabetes. Nevertheless, because of incomplete documentation of the family history of diabetes on medical records, we could not identify the effects of a family history of diabetes, whether patrilineal or matrilineal, on GDM. However, our findings show significant associations between a family history of diabetes and GDM in Taiwan. Several studies have suggested that women with GDM appear to carry susceptibility genes for type 2 diabetes (Wung and Lin, 2011). Health care providers should highlight the importance of nutrition, exercise, and lifestyle during pregnancy among women with a family history of diabetes. The results of the univariate logistic regression analysis suggest that multipara were at a high risk of GDM in Taiwan. However, after adjustment for other factors, there were no significant differences in parity. This finding is similar to that reported by Dode and Santos (2009). Similarly, the other studies have derived negative correlations between body height and the risk of GDM (Ogonowski and Maizgowski, 2010; Rudra et al., 2007). However, because body height may be affected by genes, nutrition, hormones, and socioeconomic status, it is not considered a typical risk
19
factor for GDM. In the current study, shorter Taiwanese women appeared to have a high risk of GDM; however, there were no significant differences in the association of body height with GDM after adjustment for other factors. The prepregnancy BMI was a critical factor for GDM in Taiwan. Previous studies have supported that a higher prepregnancy BMI is associated with an increased risk of GDM (Davenport et al., 2010; van Leeuwen et al., 2010). Moreover, Morisset et al. (2010) reviewed studies published between 1975 and 2009 and confirmed the association between prepregnancy BMI and GDM. The prepregnancy BMI reflects the woman's nutritional status prior to pregnancy. Radeskya et al. (2008) concluded that the nutritional status prior to pregnancy was more crucial than that during pregnancy in the development of GDM. This study used a retrospective chart review design to collect data; therefore, it was not possible to obtain further details regarding body weight gain in women during the period from pregnancy to being diagnosed with GDM. Therefore, we did not analyse the association of GDM with the maternal body weight gain during pregnancy and before diagnosis of GDM. Moreover, the external validity of this study is limited because the sample was drawn from only one medical centre. Nurses should be educated and informed about the risk factors for GDM as well as the effects of GDM on maternal and child health. Moreover, health care providers should not only promote the importance of GDM screening to pregnant women but also provide perinatal care to women with GDM, thereby facilitating maternal and child health.
Conclusions Similar to other countries, the risk factors for GDM in Taiwan included advanced maternal age, a low education level, family history of diabetes, and high prepregnancy BMI. Among all the risk factors, BMI is the only intervened variable. Therefore, health care providers should proactively promote the importance of maintaining an ideal BMI to childbearing women in addition to performing risk assessment for GDM. Future research should focus on accumulating more information about the effectiveness of early intervention for pregnant women with a high risk of GDM.
Author contributions Conceived and designed the experiments: PCL. Performed the experiments: PCL. Analysed the data: PCL. Contributed reagents/ materials/analysis tools: TFC KCL HYY YLT. Wrote the paper: PCL CHH.
Author disclosure statement No competing financial interests exist.
Acknowledgements This study was funded in part by the National Science Council, Taiwan (NSC 101-2314-B-037-024-MY3).
20
P.-C. Lin et al. / Midwifery 42 (2016) 16–20
References Anna, V., van der Ploeg, H.P., Cheung, N.W., Huxley, R.R., Bauman, A.E., 2008. Sociodemographic correlates of the increasing trend in prevalence of gestational diabetes mellitus in a large population of women between 1995 and 2005. Diabetes Care 31, 2288–2293. Carolan, M., Davey, M.A., Biro, M.A., Kealy, M., 2012. Maternal age, ethnicity and gestational diabetes mellitus. Midwifery 28, 778–783. Chen, Y., Quick, W.W., Yang, W., et al., 2009. Cost of gestational diabetes mellitus in the United States in 2007. Population Health Management 12, 165–174. Davenport, M.H., Campbell, M.K., Mottola, M.F., 2010. Increased incidence of glucose disorders during pregnancy is not explained by pre-pregnancy obesity in London, Canada. BMC Pregnancy and Childbirth 10, 85. Dode, M.A.S.O., Santos, I.S., 2009. Non classical risk factors for gestational diabetes mellitus: a systematic review of the literature. Cadernos de Saude Publica 25, S341–S359. Galtier, F., 2010. Definition, epidemiology, risk factors. Diabetes & Metabolism 36, 628–651. Golden, S.H., Bennett, W.L., Baptist-Roberts, K., et al., 2009. Antepartum glucose tolerance test results as predictors of type 2 diabetes mellitus in women with a history of gestational diabetes mellitus: a systematic review. Gender Medicine 6, 109–122. Hedderson, M.M., Ferrara, A., 2008. High blood pressure before and during early pregnancy is associated with an increased risk of gestational diabetes mellitus. Diabetes Care 31, 2362–2367. Hsieh, F.Y., Block, D.A., Larsen, M.D., 1998. A simple method of sample size calculation for linear and logistic regression. Statistics in Medicine 17, 1623–1634. Hunsberger, M., Rosenberg, K.D., Donatelle, R.J., 2010. Racial/ethnic disparities in gestational diabetes mellitus: Findings from a population-based survey. Women’s Health Issues 20, 323–328. Khan, R., Ali, K., Khan, Z., 2013. Socio-demographic risk factors of gestational diabetes mellitus. Pakistan Journal of Medical Sciences 29, 843–846. King, K.B., Gerich, J.E., Guzick, D.S., King, K.U., McDermott, M.P., 2009. Is a history of gestational diabetes related to risk factors for coronary heart disease? Res. Nurs. Health 32, 298–306. Koch, E., Thorp, J., Bravo, M., Gatica, S., Romero, C.X., Aguilera, H., Ahlers, I., 2012. Women's education level, maternal health facilities, abortion legislation and maternal deaths: a natural experiment in Chile from 1957 to 2007. PLoS One 7, e36613. Kuti, M.A., Abbiyesuku, F.M., Akinlade, K.S., et al., 2011. Oral glucose tolerance testing outcomes among women at high risk for gestational diabetes mellitus.
Journal of Clinical Pathology 64, 718–721. Lin, C.H., Wen, S.F., Wu, Y.H., Huang, M.J., 2009. Using the 100-g oral glucose tolerance test to predict fetal and maternal outcomes in women with gestational diabetes mellitus. Chang Gung Medical Journal 32, 283–289. Mitanchez, D., 2010. Foetal and neonatal complications in gestational diabetes: perinatal mortality, congenital malformations, macrosomia, shoulder dystocia, birth injuries, neonatal complications. Diabetes & Metabolism 36, 617–627. Morisset, A.S., St-Yves, A., Veillette, J., Weisnagel, S.J., Tchernof, A., Robitaille, J., 2010. Prevention of gestational diabetes mellitus: a review of studies on weight management. Diabetes Metabolism Research and Reviews 26, 17–25. Ogonowski, J., Miazgowski, T., 2010. Are short women at risk for gestational diabetes mellitus? European Journal of Endocrinology 162, 491–497. Ogonowski, J., Miazgowski, T., Engel, K., Celewicz, Z., 2014. Birth weight predicts the risk of gestational diabetes mellitus and pregravid obesity. Nutrition 30, 39–43. Parmenter, K., Waller, J., Wardle, J., 2009. Demographic variation in nutrition knowledge in England. Health Education Research 15, 163–174. Radeskya, J.S., Oken, E., Rifas-Shiman, S.L., Kleinman, K.P., Rich-Edwards, J.W., Gillman, M.W., 2008. Diet during early pregnancy and development of gestational diabetes. Paediatric and Perinatal Epidemiology 22, 47–59. Rhee, S.Y., Kim, J.Y., Woo, J.T., Kim, Y.S., Kim, S.H., 2010. Familial clustering of type 2 diabetes in Korean women with gestational diabetes mellitus. Korean Journal of Internal Medicine 25, 269–272. Rudra, C.B., Sorensen, T.K., Leisenring, W.M., Dashow, E., Williams, M.A., 2007. Weight characteristics and height in relation to risk of gestational diabetes mellitus. American Journal of Epidemiology 165, 302–308. Ryan, E.A., 2013. Clinical diagnosis of gestational diabetes. Clinical Obstetrics & Gynaecology 56, 774–787. Savitz, D.A., Janevic, T.M., Engel, S.M., Kaufman, J.S., Herring, A.H., 2008. Ethnicity and gestational diabetes in New York City, 1995-2003. BJOG 115, 1969-978. van der Ploeg, H.P., van Poppel, M.N.M., Chey, T., Bauman, A.E., Brown, W.J., 2011. The role of pre-pregnancy physical activity and sedentary behaviour in the development of gestational diabetes mellitus. Journal of Science and Medicine in Sport 14, 149–152. van Leeuwen, M., Opmeer, B.C., Zweers, E.J., et al., 2010. Estimating the risk of gestational diabetes mellitus: a clinical prediction model based on patient characteristics and medical history. BJOG 117, 69–75. Wung, S.F., Lin, P.C., 2011. Shared genomics of type 2 and gestational diabetes mellitus. Annual Review of Nursing Research 29, 227–260. Yang, H., Wei, Y., Gao, X., et al., 2009. Risk factors for gestational diabetes mellitus in Chinese women-a prospective study of 16286 pregnant women in China. Diabetic Medicine 26, 1099–1104.
Contents lists available at ScienceDirect
Midwifery journal homepage: www.elsevier.com/midw
The risk factors for gestational diabetes mellitus: A retrospective study Pei-Chao Lin, PhD, RN (Assistant professor)a, Chich-Hsiu Hung, PhD, RN (Professor)a,n, Te-Fu Chan, PhD, MD (Associate professor)b,c, Kuan-Chia Lin, PhD (Professor)d, Yu-Yun Hsu, PhD, RN (Associate professor)e, Ya-Ling Tzeng, PhD, RN (Professor & Director)f a
School of Nursing, Kaohsiung Medical University, No. 100, Shih-Chuan 1st Rd., Kaohsiung City 80708, Taiwan Department of Obstetrics and Gynecology, School of Medicine, Kaohsiung Medical University, No. 100, Shih-Chuan 1st Rd., Kaohsiung City 80708, Taiwan c Department of Obstetrics and Gynecology, Kaohsiung Medical University Chung-Ho Memorial Hospital, No. 100, Tz-You 1st Rd., Kaohsiung City 80756, Taiwan d Institute of Hospital and Health Care Administration, National Yang-Ming University, No. 155, Linong Street Sec. 2, Taipei City 11221, Taiwan e Department of Nursing, College of Medicine, National Cheng Kung University, 1 University Road, Tainan City, 70101, Taiwan f School of Nursing and Graduate Institute of Nursing, China Medical University, No. 91, Hsueh-Shih Rd., Taichung City 40402, Taiwan b
art ic l e i nf o
a b s t r a c t
Article history: Received 25 January 2016 Received in revised form 15 July 2016 Accepted 21 September 2016
Objective: To investigate the risk factors for developing GDM among Taiwanese pregnant women. Design: A retrospective cohort and case-control study. Setting: At a medical centre in Southern Taiwan. Participants: The hospitalised pregnant women who were diagnosed with either GDM or normal glucose tolerance (NGT) between 1997 and 2011. The glucose tolerance test results were interpreted according to criteria established by the National Diabetes Data Group for GDM. Participants were divided into either a GDM group (case group) or a normal glucose tolerance (NGT) group (control group) in order to determine the risk factors for GDM. Measurements: With a retrospective chart review, data regarding demographics, a family history of diabetes, history of gestation, and physiological index for pre- and postpregnancy periods were collected. χ2 tests and independent t tests were used to examine the correlations between demographic characteristics and GDM. Stepwise multivariate logistic regression was used to determine the factors associated with GDM. Findings: The results of the comparison between the GDM group (n ¼ 106) and the NGT group (n ¼ 406) showed that the risk factors for GDM were maternal age, education, a family history of diabetes, and prepregnancy body mass index (BMI). Key conclusion and implication for practice: Older age, lower levels of education, a family history of diabetes, and higher prepregnancy BMI were significant risk factors for GDM. In addition to performing risk factor assessment, health care providers should proactively promote the importance of GDM screening to pregnant women at their first antenatal visit. & 2016 Elsevier Ltd. All rights reserved.
Keywords: Age Body mass index Education status Gestational diabetes mellitus Taiwan
Introduction Gestational diabetes mellitus (GDM) is a major global public health concern, with prevalence increasing yearly. Women with diabetes have high risk to give birth to giant babies and to suffer from mortality (Ryan, 2013). GDM can result in serious maternal, foetal, and neonatal health implications, including preeclampsia,
n
Corresponding author. E-mail addresses: [email protected] (P.-C. Lin), [email protected] (C.-H. Hung), [email protected] (T.-F. Chan), [email protected] (K.-C. Lin), [email protected] (Y.-Y. Hsu), [email protected] (Ya-Ling Tzeng). http://dx.doi.org/10.1016/j.midw.2016.09.008 0266-6138/& 2016 Elsevier Ltd. All rights reserved.
shoulder dystocia, macrosomia, neonatal hypoglycaemia, etc (Mitanchez, 2010). The reported prevalence of GDM worldwide ranges from 2% to 6%; the prevalence of GDM in India, the Middle East, and Sardinia ranges from 10% to 22% (Galtier, 2010). Asian women appear to be at a higher risk of developing GDM than non-Hispanic white women (Savitz et al., 2008). The prevalence of GDM in China was 4.3% (Yang et al., 2009), and a recent study reported a GDM prevalence of up to 7.4% in Taiwan (Lin et al., 2009). NonHispanic Asian/Pacific Islander women were 2.26-fold more likely to develop GDM compared with non-Hispanic white women (Hunsberger et al., 2010). GDM not only adversely affects maternal and child health but also increases the resulting medical costs (Chen et al., 2009). Several studies have reported that women with
P.-C. Lin et al. / Midwifery 42 (2016) 16–20
a history of GDM are at an increased risk of type 2 diabetes (Golden et al., 2009) and cardiovascular diseases (King et al., 2009). Therefore, health care providers should understand the risk factors for GDM and provide pregnant women with preventive care and early intervention. Maternal age is widely considered one of the risk factors associated with GDM (Dode and Santos, 2009). The risk of developing GDM increases with maternal age and an increasing prepregnancy body mass index (BMI) (Dode and Santos, 2009). The results of a study demonstrated that women with lower levels of education had a higher risk of developing GDM than that of women with higher levels of education (van der Ploeg et al., 2011). However, another study found no association between GDM and education in Chinese pregnant women (Yang et al., 2009). Numerous studies have demonstrated that women with a family history of diabetes had a higher risk of developing GDM than that of those without (van Leeuwen et al., 2010; Rhee et al., 2010). Dode and Santos (2009) reviewed relevant literature and discovered that, of the 23 articles addressing the associations between GDM and parity, only five present positive associations. Most studies have indicated that a short stature was positively associated with GDM (Dode and Santos, 2009); however, we cannot rule out the possibility of selection bias on these studies. Despite the high prevalence of GDM in Asia, few studies have focused on the risk factors for GDM, and information about factors associated with GDM among the Asian population is limited. Additional research is required to validate previous research on the factors associated with GDM. Therefore, the purpose of this study was to investigate the risk factors for developing GDM among Taiwanese pregnant women.
Methods Design This study used a retrospective chart review to collect data. The glucose tolerance test results were interpreted according to criteria established by the National Diabetes Data Group for GDM. A case-control study was conducted to determine the risk factors for GDM. Participants were divided into either a GDM group (case group) or a normal glucose tolerance (NGT) group (control group). Participants and procedure In our study, the inclusive criteria were women who were diagnosed with GDM, and gave birth and were discharged from the study hospital. The exclusive criteria involving women's medical records indicated incomplete glucose tolerance test, women with a diagnosis of type 1 diabetes or type 2 diabetes before pregnant. The minimum sample size for logistic regression was calculated according to a artificial milk (Hsieh et al., 1998) with the study results of Yang et al. (2009). If the statistical power (1 β) was 0.80, α was 0.05, and the number of participants in the control group was triple for the number of participants in the case group, the minimum sample size was 123 participants in the case group and 369 participants in the control group. This study was conducted at a medical centre in Southern Taiwan. Prior to data collection, the study was reviewed and approved by the institutional review board of the study hospital (KMUHIRB-2012-03-08(I)). Data were collected from the medical records of hospitalised women who delivered in and then were discharged from the study hospital between 1997 and 2011 with the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnostic codes of 648 (n ¼188) and 650 (n ¼678). A total of 866 medical records of patients met the
17
inclusion criteria, of which 122 could not be retrieved. After reviewing the available medical records of 744 pregnant women, we excluded 197 women who lacked data on the glucose tolerance test, antenatal visits, or child delivery. Furthermore, we excluded 10 women with a diagnosis of type 1 diabetes, 14 with type 2 diabetes, and 11 with impaired glucose tolerance. Eventually, 512 patients, 106 whose first diagnosis was GDM and 406 whose first diagnosis was NGT, were included in the analysis (Fig. 1). Data analysis Descriptive statistics were used to present the demographic characteristics of the participants. χ2 tests and independent t tests were used to examine the correlations between demographic characteristics and GDM. Stepwise multivariate logistic regression was used to analyse the risk factors for GDM. IBM SPSS Statistics Version 20 (Mandarin Chinese Edition) was used for data filing and analysis.
Findings The results of comparing the GDM group (n ¼106) with the NGT group (n¼ 406) showed significant differences (po .05) in mean age, education, employment status, a family history of diabetes, parity, body height, and prepregnancy BMI (Table 1). The results of the univariate logistic regression analysis indicated that the risk of developing GDM increased by 13% (95% confidence interval [CI]: 1.07–1.19, p o.001) for every year of maternal age, that women with an education level of junior college or below had a 3.28-fold higher risk (95% CI: 2.11–5.10, po .001) of GDM than that of women with an education level of university or above, and that unemployed women had a 1.93-fold higher risk (95% CI: 1.25–2.99, p¼ .003) of GDM than that of employed women. Additionally, women with a family history of diabetes exhibited a 7.16-fold higher risk (95% CI: 4.50–11.41, po .001) of GDM than that of women without a family history of diabetes; multipara had a 1.93-fold higher risk (95% CI: 1.24–3.02, p ¼ .004) of GDM than that of primipara. The risk of GDM was reduced by 5% (95% CI: 0.91–0.99, p ¼.008) for every 1-cm increase in body height; and the risk of GDM increased by 35% (95% CI: 1.26–1.46, po .001) for every 1 kg/m2 increase in prepregnancy BMI (Table 2). The results of the stepwise multivariate logistic regression analysis indicated that only four variables (age, education, a family history of diabetes, and prepregnancy BMI) were entered into the model. After adjusting for other variables, we determined that the risk of developing GDM increased by 10% (95% CI: 1.04–1.17, p¼ .002) for every year of maternal age; women with an education level of junior college or below had a 3.59-fold higher risk (95% CI: 2.07–6.21, p o.001) of GDM than that of women with an education level of university or above; women with a family history of diabetes showed a 6.79-fold higher risk (95% CI: 3.92–11.76, p o.001) of GDM than that of women without; and the risk of GDM increased by 29% (95% CI: 1.19–1.41, p o.001) for every 1 kg/m2 increase in prepregnancy BMI. To summarise, age, education, a family history of diabetes, and prepregnancy BMI were significant predictors of GDM (Table 3).
Discussion The results indicate that Taiwanese women's age, education, family history of diabetes, and prepregnancy BMI were risk factors for developing GDM. Evidence-based data reported suggested that older maternal age increased a woman's risk of developing GDM and this increase was most evident among women born in Asia
18
P.-C. Lin et al. / Midwifery 42 (2016) 16–20
A total of 866 hospitalized women who delivered in and were discharged from the study hospital between 1997 and 2011 with the ICD-9-CM diagnostic codes of 648 (n = 188) and 650 (n = 678)
343 women were excluded: 1. 122 medical records could not be retrieved 2. 10 medical records showed women with a diagnosis of type 1 DM before pregnant 3. 14 medical records indicated women with a diagnosis of Type II DM before pregnant 4. 197 medical records did not show women with any data for the glucose tolerance test, prenatal visits, or child delivery
523 women were undertaken glucose screening test 11 women were excluded due to impaired glucose tolerance 512 women included in the study
106 women with GDM
406 women with NGT Fig. 1. Flow chart of study participants.
Table 1 Demographic Characteristics of Women with and without GDM (n ¼ 512). Demographic characteristics
Age (year) Education Junior college or above Senior high or below Employment status Yes No Family history of diabetes Yes No Parity 0 1–3 Body height (cm) Prepregnancy BMI
Women with GDM mean7 SD n (%)
Women with NGT mean 7SD n (%)
T value or χ2 value
P-value
32.2 7 5.0
30.0 7 3.9
4.21 40.42
o .001 o .001
59(55.7)
342(84.2)
47(44.3)
64(15.8)
57(53.8) 49(46.2)
281(69.2) 125(30.8)
66(62.3) 40(37.7)
76(18.7) 330(81.3)
62(58.5) 44(41.5) 158.8 7 5.7 24.4 74.6
297(73.2) 109(28.8) 160.3 75.1 20.2 7 2.6
8.93
.003
79.52
o .001
8.62
.003
2.69 7.57
.007 o .001
and the Middle East (Carolan et al., 2012). The current study also demonstrated that older maternal age was associated with an increased risk of developing GDM. Therefore, health care providers should not only identify pregnant women with advanced maternal age as a high-risk group and provide them with related preventive health intervention but also proactively promote awareness of GDM and encourage older pregnant women to receive GDM
Table 2 Univariate logistic regression analysis for factors associated with GDM (n¼ 512). Demographic characteristics
OR
95% CI
P-value
Age Education, n (%) Junior college or below University or above Employment status Yes No Family history of diabetes Yes No Parity 2–3 1 Body height Prepregnancy BMI
1.13
(1.07–1.19)
o .001
3.28 1.00
(2.11–5.10)
o .001
(1.25–2.99)
.003
7.16 1.00
(4.50–11.41)
o .001
1.93 1.00 0.95 1.35
(1.24–3.02)
.004
(0.91–0.99) (1.26–1.46)
.008 o .001
1.00 1.93
screenings. Taiwanese women with an education level of junior college or below had a 3.59-fold higher risk of GDM than that of women with an education level of university or above. This finding is similar to those reported by previous studies, demonstrating that women with a low education level had an increased risk of GDM compared with women with a high education level (Hedderson and Ferrara, 2008; van der Ploeg et al., 2011). However, Khan et al. (2013) suggested that education level is not associated with GDM. Although sufficient evidence-based information about the association between GDM and education is scant, education appears to
P.-C. Lin et al. / Midwifery 42 (2016) 16–20
Table 3 Stepwise multivariate logistic regression analysis for factors associated with GDM (n¼ 512). Demographic characteristics
Adjusted OR
95% CI
P-value
Age Education Junior college or below University or above Family history of diabetes Yes No Prepregnancy BMI
1.10
(1.04–1.17)
.002
3.59 1.00
(2.07–6.21)
o.001
6.79 1.00 1.29
(3.92–11.76)
o.001
(1.19–1.41)
o.001
have a favourable impact on women's health conditions (Koch et al., 2013). A low education level may adversely affect a person's nutrition knowledge (Parmenter et al., 2009). For women with a low level of education who are at risk of developing GDM, health care providers should provide nutrition knowledge during pregnancy, thereby helping pregnant women maintain proper nutrition and ensure maternal and child health. The results of the univariate logistic regression analysis suggest that Taiwanese unemployed women were at high risk of GDM. This finding is consistent with that of Khan et al. (2013). Few studies have addressed the association between employment status and GDM. However, a previous study discovered that women with a lower socioeconomic status had an increased risk of GDM (Anna et al., 2008). In the current study, after adjustment for other variables including age, education, family history, and prepregnancy BMI, there were no significant differences in the association of employment status with GDM. Nevertheless, based on previous study findings, we suggest that health care providers pay particular attention to health risk assessment and management during pregnancy among women with low socioeconomic status. Among all the risk factors identified in our study, a family history of diabetes was the strongest predictor of GDM. Our finding regarding the risk ratio of a family history of diabetes for GDM is congruent with those reported in previous Asian studies (Rhee et al., 2010; van Leeuwen et al., 2010; Yang et al., 2011). After adjustment for other risk factors, the current study derived a higher risk ratio than those of previous Asian studies. A family history of diabetes was the major risk factor for GDM in Taiwan (Kuti et al., 2011). Additionally, Ogonowski et al. (2014) suggested that women with a matrilineal family history of diabetes were at a higher risk of GDM than women with a patrilineal family history of diabetes. Nevertheless, because of incomplete documentation of the family history of diabetes on medical records, we could not identify the effects of a family history of diabetes, whether patrilineal or matrilineal, on GDM. However, our findings show significant associations between a family history of diabetes and GDM in Taiwan. Several studies have suggested that women with GDM appear to carry susceptibility genes for type 2 diabetes (Wung and Lin, 2011). Health care providers should highlight the importance of nutrition, exercise, and lifestyle during pregnancy among women with a family history of diabetes. The results of the univariate logistic regression analysis suggest that multipara were at a high risk of GDM in Taiwan. However, after adjustment for other factors, there were no significant differences in parity. This finding is similar to that reported by Dode and Santos (2009). Similarly, the other studies have derived negative correlations between body height and the risk of GDM (Ogonowski and Maizgowski, 2010; Rudra et al., 2007). However, because body height may be affected by genes, nutrition, hormones, and socioeconomic status, it is not considered a typical risk
19
factor for GDM. In the current study, shorter Taiwanese women appeared to have a high risk of GDM; however, there were no significant differences in the association of body height with GDM after adjustment for other factors. The prepregnancy BMI was a critical factor for GDM in Taiwan. Previous studies have supported that a higher prepregnancy BMI is associated with an increased risk of GDM (Davenport et al., 2010; van Leeuwen et al., 2010). Moreover, Morisset et al. (2010) reviewed studies published between 1975 and 2009 and confirmed the association between prepregnancy BMI and GDM. The prepregnancy BMI reflects the woman's nutritional status prior to pregnancy. Radeskya et al. (2008) concluded that the nutritional status prior to pregnancy was more crucial than that during pregnancy in the development of GDM. This study used a retrospective chart review design to collect data; therefore, it was not possible to obtain further details regarding body weight gain in women during the period from pregnancy to being diagnosed with GDM. Therefore, we did not analyse the association of GDM with the maternal body weight gain during pregnancy and before diagnosis of GDM. Moreover, the external validity of this study is limited because the sample was drawn from only one medical centre. Nurses should be educated and informed about the risk factors for GDM as well as the effects of GDM on maternal and child health. Moreover, health care providers should not only promote the importance of GDM screening to pregnant women but also provide perinatal care to women with GDM, thereby facilitating maternal and child health.
Conclusions Similar to other countries, the risk factors for GDM in Taiwan included advanced maternal age, a low education level, family history of diabetes, and high prepregnancy BMI. Among all the risk factors, BMI is the only intervened variable. Therefore, health care providers should proactively promote the importance of maintaining an ideal BMI to childbearing women in addition to performing risk assessment for GDM. Future research should focus on accumulating more information about the effectiveness of early intervention for pregnant women with a high risk of GDM.
Author contributions Conceived and designed the experiments: PCL. Performed the experiments: PCL. Analysed the data: PCL. Contributed reagents/ materials/analysis tools: TFC KCL HYY YLT. Wrote the paper: PCL CHH.
Author disclosure statement No competing financial interests exist.
Acknowledgements This study was funded in part by the National Science Council, Taiwan (NSC 101-2314-B-037-024-MY3).
20
P.-C. Lin et al. / Midwifery 42 (2016) 16–20
References Anna, V., van der Ploeg, H.P., Cheung, N.W., Huxley, R.R., Bauman, A.E., 2008. Sociodemographic correlates of the increasing trend in prevalence of gestational diabetes mellitus in a large population of women between 1995 and 2005. Diabetes Care 31, 2288–2293. Carolan, M., Davey, M.A., Biro, M.A., Kealy, M., 2012. Maternal age, ethnicity and gestational diabetes mellitus. Midwifery 28, 778–783. Chen, Y., Quick, W.W., Yang, W., et al., 2009. Cost of gestational diabetes mellitus in the United States in 2007. Population Health Management 12, 165–174. Davenport, M.H., Campbell, M.K., Mottola, M.F., 2010. Increased incidence of glucose disorders during pregnancy is not explained by pre-pregnancy obesity in London, Canada. BMC Pregnancy and Childbirth 10, 85. Dode, M.A.S.O., Santos, I.S., 2009. Non classical risk factors for gestational diabetes mellitus: a systematic review of the literature. Cadernos de Saude Publica 25, S341–S359. Galtier, F., 2010. Definition, epidemiology, risk factors. Diabetes & Metabolism 36, 628–651. Golden, S.H., Bennett, W.L., Baptist-Roberts, K., et al., 2009. Antepartum glucose tolerance test results as predictors of type 2 diabetes mellitus in women with a history of gestational diabetes mellitus: a systematic review. Gender Medicine 6, 109–122. Hedderson, M.M., Ferrara, A., 2008. High blood pressure before and during early pregnancy is associated with an increased risk of gestational diabetes mellitus. Diabetes Care 31, 2362–2367. Hsieh, F.Y., Block, D.A., Larsen, M.D., 1998. A simple method of sample size calculation for linear and logistic regression. Statistics in Medicine 17, 1623–1634. Hunsberger, M., Rosenberg, K.D., Donatelle, R.J., 2010. Racial/ethnic disparities in gestational diabetes mellitus: Findings from a population-based survey. Women’s Health Issues 20, 323–328. Khan, R., Ali, K., Khan, Z., 2013. Socio-demographic risk factors of gestational diabetes mellitus. Pakistan Journal of Medical Sciences 29, 843–846. King, K.B., Gerich, J.E., Guzick, D.S., King, K.U., McDermott, M.P., 2009. Is a history of gestational diabetes related to risk factors for coronary heart disease? Res. Nurs. Health 32, 298–306. Koch, E., Thorp, J., Bravo, M., Gatica, S., Romero, C.X., Aguilera, H., Ahlers, I., 2012. Women's education level, maternal health facilities, abortion legislation and maternal deaths: a natural experiment in Chile from 1957 to 2007. PLoS One 7, e36613. Kuti, M.A., Abbiyesuku, F.M., Akinlade, K.S., et al., 2011. Oral glucose tolerance testing outcomes among women at high risk for gestational diabetes mellitus.
Journal of Clinical Pathology 64, 718–721. Lin, C.H., Wen, S.F., Wu, Y.H., Huang, M.J., 2009. Using the 100-g oral glucose tolerance test to predict fetal and maternal outcomes in women with gestational diabetes mellitus. Chang Gung Medical Journal 32, 283–289. Mitanchez, D., 2010. Foetal and neonatal complications in gestational diabetes: perinatal mortality, congenital malformations, macrosomia, shoulder dystocia, birth injuries, neonatal complications. Diabetes & Metabolism 36, 617–627. Morisset, A.S., St-Yves, A., Veillette, J., Weisnagel, S.J., Tchernof, A., Robitaille, J., 2010. Prevention of gestational diabetes mellitus: a review of studies on weight management. Diabetes Metabolism Research and Reviews 26, 17–25. Ogonowski, J., Miazgowski, T., 2010. Are short women at risk for gestational diabetes mellitus? European Journal of Endocrinology 162, 491–497. Ogonowski, J., Miazgowski, T., Engel, K., Celewicz, Z., 2014. Birth weight predicts the risk of gestational diabetes mellitus and pregravid obesity. Nutrition 30, 39–43. Parmenter, K., Waller, J., Wardle, J., 2009. Demographic variation in nutrition knowledge in England. Health Education Research 15, 163–174. Radeskya, J.S., Oken, E., Rifas-Shiman, S.L., Kleinman, K.P., Rich-Edwards, J.W., Gillman, M.W., 2008. Diet during early pregnancy and development of gestational diabetes. Paediatric and Perinatal Epidemiology 22, 47–59. Rhee, S.Y., Kim, J.Y., Woo, J.T., Kim, Y.S., Kim, S.H., 2010. Familial clustering of type 2 diabetes in Korean women with gestational diabetes mellitus. Korean Journal of Internal Medicine 25, 269–272. Rudra, C.B., Sorensen, T.K., Leisenring, W.M., Dashow, E., Williams, M.A., 2007. Weight characteristics and height in relation to risk of gestational diabetes mellitus. American Journal of Epidemiology 165, 302–308. Ryan, E.A., 2013. Clinical diagnosis of gestational diabetes. Clinical Obstetrics & Gynaecology 56, 774–787. Savitz, D.A., Janevic, T.M., Engel, S.M., Kaufman, J.S., Herring, A.H., 2008. Ethnicity and gestational diabetes in New York City, 1995-2003. BJOG 115, 1969-978. van der Ploeg, H.P., van Poppel, M.N.M., Chey, T., Bauman, A.E., Brown, W.J., 2011. The role of pre-pregnancy physical activity and sedentary behaviour in the development of gestational diabetes mellitus. Journal of Science and Medicine in Sport 14, 149–152. van Leeuwen, M., Opmeer, B.C., Zweers, E.J., et al., 2010. Estimating the risk of gestational diabetes mellitus: a clinical prediction model based on patient characteristics and medical history. BJOG 117, 69–75. Wung, S.F., Lin, P.C., 2011. Shared genomics of type 2 and gestational diabetes mellitus. Annual Review of Nursing Research 29, 227–260. Yang, H., Wei, Y., Gao, X., et al., 2009. Risk factors for gestational diabetes mellitus in Chinese women-a prospective study of 16286 pregnant women in China. Diabetic Medicine 26, 1099–1104.