|Year : 2017 | Volume
| Issue : 2 | Page : 77-82
Grandmultiparity: Is it really an independent predictor of adverse pregnancy outcomes?
M Ikeanyi Eugene, O Addah Abedinego
Department of Obstetrics and Gynecology, Niger Delta University, Wilberforce Island, Bayelsa State, Nigeria
|Date of Web Publication||15-Sep-2017|
M Ikeanyi Eugene
Department of Obstetrics and Gynecology, Niger Delta University, Wilberforce Island, P.M.B. 071, Bayelsa State
Source of Support: None, Conflict of Interest: None
Context: Grandmultiparity has been associated with increased adverse maternal and perinatal outcomes mostly in poor-resource settings. The question remains to what extent parity on itself contributes to unfavorable obstetric outcomes. There are limited studies that control for other predictor variables of maternal and perinatal outcomes in high parity pregnancies. Aim: This compared the adverse maternal and perinatal outcomes among grand multiparas with age-matched low parous counterparts to generate data for focused obstetric care. Subjects and Methods: An age-matched case–control study was performed on 2009–2014 data of all the high parity births in a second-tier mission hospital in a state capital city, Nigeria. Selected maternal and fetal outcome variables were compared between the women of high parity (para ≥5) and low parity (para 1–4). The primary outcomes were adverse maternal and perinatal variables. The statistical tests were done with statistical significance at <0.05. Results: The prevalence of grandmultipara was 2.52%, grand multiparity was significantly associated with lower educational attainment (odds ratio [OR] = 4.3, 95% confidence interval [CI]: 2.4–7.6), lower socioeconomic class (OR = 3.78, 95% CI: 2.28–6.28) and abnormal lie (OR = 10.48, 95% CI: 1.21–91.00), no significant association (P > 0.05) with stillbirths, low birth weight, macrosomia, preterm birth, chronic hypertension, gestational hypertension anemia, induction of labor, cephalopelvic disproportion/obstructed labor, and cesarean section. Conclusions: The prevalence of grandmultiparity was low at 2.52%. With the exception of abnormal lie, high parity was not associated with increased adverse fetal and maternal outcomes in this study.
Keywords: Adverse outcomes, grandmultiparity, independent, maternal, perinatal, predictor
|How to cite this article:|
Eugene M I, Abedinego O A. Grandmultiparity: Is it really an independent predictor of adverse pregnancy outcomes?. Saudi J Health Sci 2017;6:77-82
| Introduction|| |
The literature is replete with reports of an increase mean parity in most developing countries. Parity is the number of previous deliveries after fetal age of viability of ≥24 weeks irrespective of whether stillbirths or live births. However, 28 completed weeks is still the cutoff point of fetal viability in most developing countries like Nigeria. Parity remains an important index in obstetric practice. The obstetric population is stratified using parity into nullipara, primipara, multipara, and grandmultipara and even great grandmultipara for 0, 1, ≥2, ≥5, or ≥10 previous deliveries, respectively. Parity ≤4 is considered low parity and ≥5 considered high parity or grand multiparty. The term grandmultipara was first used by Solomons in his 1934 works when he noted that grandmultipara was dangerous because maternal mortality increased steadily from the fifth childbirth. The International Federation of Gynecology and Obstetrics in 1993 defined grand multiparity as delivery of 5th–9th infant whereas women who are undergoing their 10th or more delivery are considered to be great grand multipara., The literature is full of conflicting reports on outcomes of high parity pregnancy. Some studies reported increased adverse obstetric outcomes,,,,,,,,,, some reported reduced complications  while others did not confirm high parity as an independent pregnancy risk factor.,, Furthermore, the studies in the literature that associate high parity with increased maternal and fetal adverse outcomes are diverse in their findings. For example, high parity has been variously associated with increased risk of anemia,, abnormal lie; malpresentation,, preeclampsia, postpartum hemorrhage,,,, placenta previa,,, macrosomia,, preterm birth, fetal death.,,, Yet, others reported of reduced intrapartum complications where maternal age was controlled.
The differences in maternal and child mortality by parity have been partly associated to lower health services coverage among the high parity births. It was observed that there was a trend toward lower utilization of maternal health services by mothers as the birth order increases, especially institutional delivery and skilled birth attendance. In addition, a number of confounders played out in the high parity literature. Most of such reports attributed adverse obstetric outcomes to high parity., Meanwhile, socioeconomic status, maternal age, nutritional status, standard of maternal health care, and health-seeking behavior are known to affect the obstetric performance. These factors operate at different degrees in different individual mothers and different settings in some cases in combinations to influence the obstetric outcome.
Again, methodological differences are very important on the evidence. For instance, some of the reports came from cross-sectional study that is weak in assessing causality., In developed economies with excellent obstetric facilities, standard perinatal care and wholesome health-seeking behavior, the prevailing reports indicate that obstetric complications among grandmultipara are now independently associated with progressive maternal age and not parity.
Unlike the rich-resource settings with satisfactory outcomes because these confounders are either nonexistent or reduced to an insignificant level, in poor-resource settings, these are rife with the attendant increased adverse obstetric outcomes in high parity pregnancy. The limited data from studies that controlled for these confounders in our setting informed the interest in this study designed to measure the independent contribution of high parity to adverse pregnancy outcomes with the control for maternal age.
| Subjects and Methods|| |
This age-matched case–control study took place at St Philomena Catholic Mission Hospital, situated at the center of Benin city, the Edo state capital. This is a second-tier missionary hospital of 75 years of feto–maternal services. Edo state is one of the oil rich Niger Delta regions of Southern Nigeria with the population close to 4 million. It is bounded to the North and East by Kogi state, South and West by Delta state, and Ondo state, respectively. The dominant ethnic groups are the Binis, Esan, and Afemai and high presence of residents from across other parts of the country and the world. Their main occupation is agriculture while the capital city residents are predominantly civil servants and traders. The study center serves as referral center to other health facilities within and outside the state. It has fairly well equipped and staffed obstetric, neonatal, and medical records units. Her average annual deliveries were over a thousand. The labor ward and theater delivery records between 2009 and 2014 were retrospectively reviewed to generate relevant database on the predictor variables: sociodemographic characteristics of the subjects (parity, age, booking status, education, and social class). The outcome variables measured were maternal complications; induction of labor, cesarean section rate, blood transfusion, mortality, hysterectomy, medical disorders, the neonatal complications; stillbirth, low birth weight (LBW) (a birth weight [BWT] <2500 g), macrosomia; BWT ≥4000 g, preterm birth (birth at <37 completed weeks of gestation from the 1st day of the last normal menstrual period). The variables were dichotomized for bivariate analysis. The participants in the two arms were matched for age to remove the confounding influence of age on obstetric outcomes and allow for the assessment of the independent effect of high parity on the obstetric outcome. Parity refers to the number of previous deliveries after fetal viability of ≥24 weeks of gestation irrespective of live or stillbirth. The subjects were selected from the records documented according to their delivery order. For this study, the subjects were grouped into high parity (grand multiparity ≥5 previous deliveries; the study group) and low parity (para 1–4 previous deliveries; the control group). For each grandmultipara selected, the next two eligible low parity participants were recruited as control. Our analyses excluded nulliparous, previous cesarean, and multiple pregnancies due to the risks inherent in them. The social stratification of the participants was based on the woman's educational level; scored 0, 1, and 2 for tertiary, secondary, and primary/nil formal education, respectively, and spouse occupation; scored 1, 2, and 3 for professional, middle level, and unskilled level, respectively. The aggregate score was the social class of each participant. For the purpose of this study, this was further scaled down to upper, middle, and lower class for Class I/II, III, and IV/V, respectively. Ethical approval was obtained from the hospital research and ethics committee. The consent from the subjects was not required as the subject's identifiers were not involved. The data analysis was done with EPI-INFO version 3.5.1 statistical software developed by the center for disease control and prevention (CDC) in Atlanta Georgia, USA and released in August 2008 and InStat statistical software developed by GraphPad software Inc., California. Using the two rows, two columns contingency table statistical tests for associations was done with Fisher's exact test as some of the cells contained <5 variables. Student's t-test was used for the comparison of the means as appropriate. The statistical significance was set at P < 0.05 and 95% confidence interval (CI) exclusion of nullity of one.
| Results|| |
A total of 3969 women had their delivery at the center in the reviewed period and 100 of them were grand multipara (study group), a high parity rate of 2.52%. The mean age of the study group was 36.2 ± 4.0 years, with the range of 27–44 years while the corresponding values for the control counterparts were 35.6 years and 25–44 years, respectively. The mean age difference of 0.6 years was not statistically significant (t = 1.22, P = 0.22). The mean parity of the groups was 5.5 ± 0.8 versus 2.3 ± 1.1 while the ranges were 5–9 and 1–4 for the study group and the control counterparts, respectively. The observed mean parity difference of 3.2 was statistically significant (t = 25.7, P ≤ 0.001). One case of grandmultipara with multiple pregnancy was excluded from further analyses.
[Table 1] compares the baseline characteristics of the study group and the control group. Compared to the control group, the study group was more than fourfold and about the same ratio more likely to attain secondary level of education or below (4.3, 2.4–7.6, P < 0.001) and to be of lower social class (3.78, 2.28–6.28, P < 0.001). The observed differences were statistically significant. Both arms were all married, and the study group though not statistically significant, compared with the control counterparts were more likely not to access prenatal care (odds ratio [OR] = 1.18, 95% CI: 0.45, 3.10).
|Table 1: Sociodemographic characteristics of participants cross-tabulated by subgroups (n=297)|
Click here to view
[Table 2] shows the comparative results of the participants' perinatal outcomes. Relative to the control group, the study group was associated with increased likelihood of stillbirth (OR = 1.52, 95% CI 0.33, 6.91) the observed difference was not statistically significant. The findings were not different for LBW (OR = 1.29, 95% CI: 0.42, 2.81), macrosomia (OR = 1.10, 95% CI: 0.53, 2.26), and preterm (OR = 1.25, 95% CI: 0.49, 3.45) Fewer neonates of the study group had Apgar score of <7 at 5 min (OR = 0.59, 95% CI: 0.16, 2.19) again the observed difference was not statistically significant. The neonates of the study group on the average weighed slightly less (t = 1.09, P = 0.28). However, the observed mean BWT difference of 76.4 g was not statistically significant.
[Table 3] compares the maternal outcomes of the study group and the control counterparts. The study group was comparatively more than fourfold associated with chronic hypertension (OR = 4.13, 95% CI: 0.74, 22.94) but less likely to suffer gestational hypertensive disorder (0.88, 0.37–2.10). The observed differences were not statistically significant. The study group was also less likely to suffer anemia in pregnancy but the observed difference was not statistically significant. Compared with the control counterparts, the study group was significantly more associated with fetal abnormal lie (OR: 10.5, 95% CI: 1.21, 91.0). Nonetheless, the study group was less likely to be associated with induction of labor (OR = 0.92, 95% CI: 0.43–2.00), cephalopelvic disproportion/obstructed labor (OR = 0.55, 95% CI: 0.18, 1.73), and cesarean section (OR = 0.89, 95% CI: 0.49, 1.61) compared to the control counterparts. Again, the observed differences were not statistically significant. One of the participants in the study group died, a maternal mortality ratio of 10.41/1000 live births for the high parity group and none for the low parity group. A participant each from each arm had emergency total abdominal hysterectomy for uncontrollable postpartum hemorrhage and similar for diabetes mellitus in pregnancy, respectively. These variables were too few for statistical comparison.
|Table 3: Maternal outcomes cross-tabulated by participants subgroups (n=297)|
Click here to view
| Discussion|| |
This age-matched case–control study investigated the independent contribution of high parity to adverse obstetric outcomes. Our results appeared to indicate that with the exception of abnormal lie, high parity appeared not to be independently significantly associated with negative maternal and neonatal outcomes.
The prevalence of grand multiparity in this data was 2.5% which was comparable to a range of 2.49%–7.5% reported from Southern Nigeria.,, The figure appeared lower than 18.8% from the Northern Nigeria. and 17%–33% from other parts of Sub-Saharan Africa.,, The prevalence of 3%–4% was reported in industrialized countries. For instance, a prevalence of 4.2% was reported from North America, 15.9% in Asia. and 38.7% in the Middle East  regions of the world. The low prevalence in our data may be because most of the clients to this center are well educated; high social class; financially comfortable enough to access the services at the center. Highly educated and less socially deprived are more likely to have fewer child births relative to their low literate and more socially deprived counterparts. The mean parity of the study group was comparable to the previous reports  and lower than another. The grandmultiparity in this population could suggestively be attributed to poor access and use of family planning services and subsequent increased unintended pregnancies. However, our data did not cover the reasons for high parity in the participants. It is pertinent to observe that older studies ascribed increased negative maternal and fetal outcomes to grand multiparity. Recent data, on the other hand, originate from a health system that has generally undergone and rapidly undergoing positive changes both in the poor and the rich-resources settings, and this trend seemed to have positively influenced the outcome observed in our data.
The grand multiparous participants appeared to be relatively of low educational level and lower social class corroborating other reports., Low educational attainments in female gender may serve a potential driver for early and prolong childbearing with consequent high parity as evidenced in our data. This indicates the need for female educational empowerment to delay age at first child bearing  and possibly the number of births. Education is one of the recognized means of widening the female gender social gap. In addition, education is expected to improve the woman's health-seeking behavior one of the covariates of obstetric outcome. A combination of educational and economic empowerment of the female folk will engender not only improved access to the best of the limited obstetric services in most low-resource settings but will limit the desire for large family size. Our results associated low literacy and socioeconomic status with increased high parity.
With the exception of abnormal lie the both arms seemed to be statistically comparable in the measured maternal and perinatal adverse outcomes. Improved and satisfactory contemporary obstetrics have been found to attenuate the high parity-related adverse pregnancy outcomes prevalent in older literatures. The obstetric service in the center was headed by a resident obstetrician and was fairly satisfactory. However, the high parity women appeared to be more prone to abnormal lie relative to control group corroborating another report. Abnormal lie and malpresentation common in high parous women may be attributable to unstable gravid uterus due to lax anterior abdominal wall and uterine muscles from repeated pregnancies and overstretching of the muscles by the enlarged gravid uterus. Similarly, increased tendency for obstetric hemorrhage as reported in the literature , may be ascribed to the same reason of lax uterine wall muscles and atony and placenta previa in high parity although our data failed to confirm this. Prenatally, the both arms of participants appeared comparable in hypertensive disorders in pregnancy. Nevertheless, the study group appeared insignificantly more prone to chronic hypertensive disorder as in another report  and less prone to gestational hypertensive disorders contrary to reports by other researchers who noted significant increase incidence of hypertensive diseases in the grand multipara in their study.,,, Neither of these was to a significant level in our data. The data on diabetes mellitus in pregnancy, however, was too few for comparison. With the control of maternal age in this data, high parity could not be confirmed a risk factor for increased cesarean delivery reported in the literature.,, However, some of these evidences emanated from cross-sectional study weak in the proof of causality., Due to paucity of data for any comparative analytical statement, our results could not confirm the reports of increased risk of maternal death in high parity group reported by other researchers.,, Anemia in pregnancy has been associated with high parity,,,,, but our data again could not confirm these reports.
The perinatal outcomes in both the high and low parity participants in overall appeared statistically comparable corroborating another report in the literature. This seemed to exclude the parity factor as the predisposing factor for adverse neonatal outcomes in high parity pregnancies. This conclusion was notwithstanding the slight and insignificant increase in the risk of stillbirths and perinatal deaths among the high parity women. It is evident from the studies where the influence of maternal age of high parity mothers was measured and not controlled there was a significant increase of adverse obstetric outcomes.,,,, This, however, was not confirmed by Wang et al. whose data indicated that with the exception of increase in the risk of interventional delivery including cesarean section among older women, there was no significant difference in adverse obstetric and perinatal outcomes across the age and parity subgroups. In addition, our results seemed to disagree with the association of high parity with fetal macrosomia reported by other researchers.,,,
Unlike another report in the literature, our data could not significantly associate high parity with increase preterm birth in accord to another report.
One other study has shown that grand multiparous women had no increased risk of adverse pregnancy outcome where women were economically stable or had satisfactory access to health care. Kozuki et al. could not demonstrate the association of high parity with adverse neonatal outcome. They subsequently attributed the association observed between high parity and adverse outcome to confounders rather than to any biological mechanisms. This was consistent with the findings by the previous researchers. Age of the woman, the environment, and socioeconomic class have been previously variously implicated covariates for pregnancy outcome in high parity pregnancy.,, Kenny et al. reported an association of advanced maternal age with a range of adverse pregnancy outcome independent of parity and socioeconomic status. From our data, high parity appeared to have failed to demonstrate a significant independent role in increasing adverse perinatal outcomes. This appeared to corroborate other reports,, but could not confirm the findings from similar researches., High parity has been associated with maternal insulin resistance and hyperglycemia  known to be associated with increase fetal weight. In this study, the two arms of participants were comparable in their rate of birth to macrosomic neonates though the study group had more of such births. Reported association of perinatal mortality with grand multiparity  could not be confirmed by our results.
The findings of another age-matched study  indicate that age and not parity per se increases the observed negative outcomes including chronic medical disorders such as hypertensive disorders and diabetes mellitus. Aging and degenerative changes are directly related resulting in some of the chronic medical disorders. Chronic medical disorders in turn have negative effects on the conceptus via impairment of placenta function.
This study lacked the control for other demographics known to influence pregnancy outcomes. These include socioeconomic state and educational level of the subjects both known to influence the health-seeking behavior and therefore, the pregnancy outcome. This study relied on data from one center. Findings from a multicenter data are known to be more representative of the studied population. In addition, our sample size appeared small; the larger a sample size the better the power of the study. Retrospective data are fraught with inadequacies too. The study majorly derived its strength from its age-matched case–control design that allows the comparison of the measured outcome variables.
| Conclusions|| |
With the exception of abnormal lie, the participants of high and low parity in this data appeared comparable in perinatal and maternal adverse outcomes. The preponderance of evidence in this study, therefore, did not appear supportive of high parity as a significant independent pregnancy risk factor. Suggestively, confounders such as maternal age and health-seeking behavior at personal level and prevailing obstetric practices and accessibility at community level might be in interplay in grand multiparity pregnancy to heighten the adverse obstetric outcome seen in grand multiparity.
To strengthen the available evidence base on the role of high parity in pregnancy outcome, evidence from a large sample sized and possibly a multicenter survey that control for other relevant parity confounders is suggested. Until this evidence is made available, the efforts should be directed at improving the standards of obstetric services, women empowerment, and counseling on the socioeconomic implications of large family size.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Borton C. Gravidity and Parity Definitions (and Their Implications in Risk Assessment).. https:// patient info; 2009
. [Last retrieved on 2013 Jun 26].
Solomons B. The dangerous multipara. Lancet 1934;224:8-11.
Babinszki A, Kerenyi T, Torok O, Grazi V, Lapinski RH, Berkowitz RL, et al.
Perinatal outcome in grand and great-grand multiparity: Effects of parity on obstetric risk factors. Am J Obstet Gynecol 1999;181:669-74.
DieJomaoh FM, Omene JA, Omu AE, Faal MK. The problems of grandmultiparity as seen at Benin Teaching Hospital, Benin-City, Nigeria. Trop J Obstet Gynaecol 1985;5:13-7.
Al-Farsi YM, Brooks DR, Werler MM, Cabral HJ, Al-Shafei MA, Wallenburg HC, et al.
Effect of high parity on occurrence of anemia in pregnancy: A cohort study. BMC Pregnancy Childbirth 2011;11:7.
Jacqumyn Y, Senten L, Vellinqa S, Vermeulen K, Martens G. Does practice make perfect? An age-matched study on grand multiparty in Flanders, Belgium. J Perinat 2006;34:28-31.
Geidam AD, Audu BM, Oummate Z. Pregnancy outcome among grand multiparous women at the university of Maiduguri teaching hospital: A case control study. J Obstet Gynaecol 2011;31:404-8.
Mgaya AH, Massawe SN, Kidanto HL, Mgaya HN. Grand multiparity: Is it still a risk in pregnancy? BMC Pregnancy Childbirth 2013;13:241.
Rayamajhi R, Thapa M, Pande S. The challenge of grand-multiparity in obstetric practice. Kathmandu Univ Med J (KUMJ) 2006;4:70-4.
Alsammani MA, Ahmed SR. Grandmultiparity: Risk factors and outcome in a tertiary hospital: A comparative study. Med Arch 2015;69:38-41.
Nasar AH, Fayyumy R, Saab W, Mehio G, Usta IM. Grandmultiparas in modern Obstetrics. Am J Perinatol 2006;23:345-9.
Roman H, Robillard PY, Verspyck E, Hulsey TC, Marpeau L, Barau G, et al.
Obstetric and neonatal outcomes in grand multiparity. Obstet Gynecol 2004;103:1294-9.
Afolabi AF, Adeyemi AS. Grand-multiparity: Is it still an obstetric risk? Open J Obstet Gynecol 2013;3:411-5.
Rizwan N, Parveen G, Abbasi RM. Frequency of grand multiparity and its fetomaternal outcome at Liaquat University Hospital, Hyderabad. Isr Med J 2009;1:49-53.
Shah PS, Knowledge Synthesis Group on Determinants of LBW/PT births. Parity and low birth weight and preterm birth: A systematic review and meta-analyses. Acta Obstet Gynecol Scand 2010;89:862-75.
Humphry MD. Is grand multiparty an independent predictor of pregnancy risk? A retrospective observational study. Med J Aust 2003;179:294-6.
Ozkan ZS, Atılgan R, Goktolga G, Sımsek M, Sapmaz E. Impact of grandmultiparity on perinatal outcomes in Eastern region of Turkey. J Matern Fetal Neonatal Med 2013;26:1325-7.
Bugg GJ, Atwal GS, Maresh M. Grandmultiparae in a modern setting. BJOG 2002;109:249-53.
Sonneveldt E, DeCormier Plosky W, Stover J. Linking high parity and maternal and child mortality: What is the impact of lower health services coverage among higher order births? BMC Public Health 2013;13 Suppl 3:S7.
Ogedengbe OK, Ogunmokun AA. Grandmultiparity in Lagos, Nigeria. Niger Postgrad Med J 2003;10:216-9. [Full text]
Kenny LC, Lavender T, McNamee R, O'Neill SM, Mills T, Khashan AS, et al.
Advanced maternal age and adverse pregnancy outcome: Evidence from a large contemporary cohort. PLoS One 2013;8:e56583.
Olusanya O, Okpere EE, Ezimokhai M. The importance of social class in voluntary fertility control in a developing country. West Afr J Med 1985;4:205-12.
Ikeako LC, Nwajiaku L. Grandmultiparity: Experience at Awka, Nigeria. Niger J Clin Pract 2010;13:301-5.
] [Full text]
Abasiattai AM, Utuk NM, Udoma EJ, Umoh AV. Grandmultiparity: Outcome of delivery in a tertiary hospital in Southern Nigeria. Niger J Med 2011;20:345-8.
Idoko P, Nkeng G, Anyawu M. Reasons for current pregnancy amongst grand multiparous Gambian women - A cross sectional survey. BMC Pregnancy Childbirth 2016;16:217.
Yasir R, Perveen FZ, Ali L, Perveen S, Tayyab S. Grand-multiparity still an obstetric risk for developing countries. Med Channel 2010;16:264-8.
Hoque M, Hoque E, Kader SB. Pregnancy complications of Grand-multiparity at a rural setting of South Africa. Iran J Reprod Med 2008;6:25-31.
Hamilton BE, Martin JA, Osterman MJ, Curtin SC, Matthews TJ. Births: Final data for 2014. Natl Vital Stat Rep 2015;64:1-64.
Ferre C. Age at First Child-Does Education Delay Fertiliy Timing? In Policy Research Working Bank TW; 2009.
Aliyu MH, Jolly PE, Ehiri JE, Salihu HM. High parity and adverse birth outcomes: Exploring the maze. Birth 2005;32:45-59.
Wang Y, Tanbo T, Abyholm T, Henriksen T. The impact of advanced maternal age and parity on obstetric and perinatal outcomes in singleton gestations. Arch Gynecol Obstet 2011;284:31-7.
Kumari AS, Badrinath P. Extreme grandmultiparity: Is it an obstetric risk factor? Eur J Obstet Gynecol Reprod Biol 2002;101:22-5.
Kozuki N, Lee AC, Silveira MF, Sania A, Vogel JP, Adair L, et al.
The associations of parity and maternal age with small-for-gestational-age, preterm, and neonatal and infant mortality: A meta-analysis. BMC Public Health 2013;13 Suppl 3:S2.
Seidman DS, Dollberg S, Stevenson DK, Gale R. The effects of high parity and socioeconomic status on obstetric and neonatal outcome. Arch Gynecol Obstet 1991;249:119-27.
Fowler-Brown AG, de Boer IH, Catov JM, Carnethon MR, Kamineni A, Kuller LH, et al.
Parity and the association with diabetes in older women. Diabetes Care 2010;33:1778-82.
[Table 1], [Table 2], [Table 3]