Home

ISLAMIC MEDICAL EDUCATION RESOURCES 04

9812-EPIDEMIOLOGICAL STUDY OF THE DETERMINANTS OF LOW BIRTH WEIGHT IN KUANTAN DISTRICT, PAHANG, MSIA

Dr Zainal Fitri (Kuantan General Hospital) & Prof Omar Hasan Kasule (Faculty of Medicine, International Islamic University PO Box Kuantan Pahang, Dar al Makmur, Malaysia)

ABSTRACT

The objective of this case control study is to identify the major risk factors for low-birth weight in preparation for a much larger prospective population-based follow-up study. The case series were 100 women with babies of birth weight less than 2500 grams delivered at Kuantan General Hospital over a period of about 1 year, June 1996-May 1997. Two controls were selected for each case from women who delivered  normal weight babies at about the same time in the same hospital. A detailed questionnaire on socio-demographic and biological variables was administered in addition to physical examination of the mothers and babies. The major risk factors for low birth weight identified were: zip code of residence, husband's occupation as clerical and general support, ownership of a motor-bike, ante-natal care under a general practitioner, use of anti-hypertensive drugs, pre-ecclampsia, placenta previa, pyrexia in pregnancy and pre-term labor. Marriage at age below 20 years and use of  contraceptives were associated with a lower risk of low birth weight babies. A follow-up study will be undertaken in which a larger number of pregnant women will be studied to identify risk factors for low birth weight while controlling rigorously for all confounders.

 

INTRODUCTION

Low birth weight is internationally defined by World Health Organization as weight at birth below 2500 grams (1). There are babies in East Asia who are perfectly healthy but weigh below this which has raised the issue of whether the international standards are valid for a country like Malaysia where both adults and newborns are normally smaller. For the purpose of this study we have used the international standards because with urban residence and good nutrition Asian newborns achieve birth weights comparable to those in Europe and the US. Studies of the causes of low-birth weight in Malaysia (2,3,4,5,6,7) and other countries (8,9,10,11,12,13,14,15,16,17,18,19,20,21,22) indicated multiple causes of low birth weight that include socio-demographic and biological factors. Yadav (1) studied birth weights in Lundu, Sarawak, East Malaysia, and found an incidence of low birth weight of 11.84%. The risk factors found were: ethnicity, pre-term delivery,  young mothers 15-19 years, older mothers above 35 years, anemia, hypertension and pre-eclampsia. Two unpublished Malaysian studies on low-birth weight require special mention. Marlia Mohamed Salleh studied a sample of births in 1986 at the Hospital Sultanah Aminah in Johor Baru (6). The study covered 950 babies, 13.7% were low-birth weight, below 2500 grams. The risk of low birth weight varied by ethnic group: Malay (13.7%), Indians (27.5%), and Chinese (11.6%). Other risk factors identified were: young mothers aged 15-24 years, mothers aged over 35 years, primiparity, previous abortion, anemia and hypertensive disease. She found that higher levels of education were associated with lower risk. Bailah Leigh (7) studied obstetric outcome in squatter settlements in Kuala Lumpur and concluded that there was no significant relation between social deprivation and risk of low birth weight. Despite overall socio-economic development and improvement in health status over the past decade, low birth weight persists in some sectors of the Malaysian population (23). Mortality in the Very low Birth Weight (VLBW) babies is still high (2). The purpose of the present case-control hospital-based study was to identify epidemiological correlates of low birth weight to enable planning and execution of population-based follow-up and intervention studies. The special features of this study is use of a rigorous case control methodology and exploration of many potential determinants employing high-quality data.

 

METHODS

The study was carried out at the Kuantan General Hospital (Hospital Tengku Ampuan Afzan, HTAA), the major referral hospital on the East coast of peninsular Malaysia serving the state of Pahang and southern parts of the state of Terenganu. Mothers giving birth in the hospital in the period June 1996 to May 1997 were recruited into the study as described below. Those with infants weighing less than 2500 grams were all recruited as cases. Two controls were selected for each case being mothers whose babies had birth weight 2500 grams and above and who were recorded immediately after the case in the birth register. There were only 3 exclusion criteria: multiple births, non-citizens who have resided less than 10 years in Malaysia, and cases of delivery before arrival in hospital. Recruitment into the study continued until accrual of 100 cases and 200 controls. The results of international and Malaysian studies mentioned above were used to select variables investigated in this study (Table #1).  An extensive questionnaire was completed for both cases and controls by review of the charts, interview of the mothers and physical examination of both the mother and the infant. All interviews and examinations were carried out by a medical officer of the obstetrics unit who works in the hospital and is familiar with the patients. Simple univariate analysis was carried out to study the distribution of the study variables. Covariates were then tabulated in 2x2 contingency tables comparing cases and controls. Covariates significantly associated with the study outcome, birthweight, were identified and were included in logistic regression models to identify independent correlates of low birth weight. Socio-demographic and biological variables were investigated in two separate models.

 

RESULTS

Bivariate analysis: The case and control series were comparable in the distribution of the following socio-demographic covariates: ethnicity, state and district of residence, and rural-urban distribution. Tables #2 shows results of the analyses of the study data. Only variables that were significantly associated with low birth weight are reported. Cases had a higher proportion of the following socio-economic variables: zipcode of residence, husband's occupation being clerical & general and ownership of a motor-bike. Age at marriage less than 20 was associated with a lower risk. Factors of current pregnancy associated with low birth weight were: use of anti-hypertensive drugs, ecclampsia or pre-ecclampsia, heart disease, placenta previa and pyrexia. Receiving ante-natal care under  a general practitioner had a higher risk of low birth weight. The trimester when ante-natal care started as well as the number of antenatal visits had no impact on the risk of low birth weight. There were consistent differences between cases and controls during ante-natal monitoring but they did not always reach statistical significance. Women who gave birth to babies below 2.5 kg had lower body weight, higher systolic and diastolic blood pressures, lower pulse rate and smaller fundal height at follow-up visits. Cases examined in the first trimester of pregnancy had lower mean hemoglobin levels. Cases were more likely to have proteinuria in the second and third trimesters. Controls were more likely to have reported contraceptive use and a good previous obstetric outcome. Cases had higher proportions of prematurity, placenta previa and molar pregnancy in their obstetric histories. They also had worse labor experience and outcome from the current pregnancy with higher proportions of neo-natal morbidity, neonatal mortality, small for gestational age babies, and respiratory distress syndrome.

Multivariate analysis: A logistic regression model was constructed using the significant predictor variables identified in the bivariate analysis above. Two models were built one for socio-demographic and the other for biological factors to be able to study them two categories of factors separately since they act independently and need not be necessarily related (18). None of the socio-demographic variables was an independent predictor of low birth weight. This is because these variables are closely inter-related (20) and there is high collinearity among them. Two biological factors were significant independent predictors of low birth weight: use of anti-hypertensive drugs and pre-term labor.

Summary: The final results of the study are as follows: The major risk factors for low birth weight identified were: zip code of residence, husband's occupation as clerical and general support, ownership of a motor-bike, ante-natal care by a general practitioner, use of anti-hypertensive drugs, pre-ecclampsia, placenta previa, pyrexia and pre-term labor. Marriage age below 20 years and use of contraceptives were associated with a lower risk of low birth weight.

 

DISCUSSION

The study has shown that several socio-economic and biological factors interact to cause low birth weight. Prevention of low birth weight requires a multi-dimensional approach (Hughes et al 1995, Chomitz et al 1995) involving socio-economic improvement and specific biological interventions. This study has confirmed as well as contradicted findings previously reported in the literature. It has described new and unique relationships that are worth exploring further. It is not surprising that there are inconsistencies in findings of various studies because the scales of measurement of socio-economic variables are not very specific and the biological relationships involved are complex. Residence in zip code 25150 compared to other zip codes was associated with higher risk of low birth weight (crude odds ratio, OR=2.2, p=0.052). Further studies are needed to identify neighborhood factors that could explain this. Husbands working as clerks and general assistants had wives with a higher risk of low birth weight  compared to husbands in the rest of job categories investigated (crude OR=3.2, p=0.001) which is a surprising finding in view of the fact that the supposedly poorer husbands doing manual work had wives whose risk was lower than this. We may speculate that there are aspects of life-style that are not income-dependent which explain this finding. The finding that women from families that own a motor-bike had a higher risk of low birth weight was also surprising (crude OR=1.4, p=0.03). Two speculative explanations could be offered for this finding: (a) riding a motor-bike by the pregnant woman may physically have an impact on placental functions. It may be postulated that vibrations and other violent movements of the motor-bike in early pregnancy may interfere with placentation and could cause focal placental separation and bleeding that lead to a degree of placental insufficiency. (b) owning a motor-bike may be a proxy indicator for social stress, poverty or lifestyle that may affect maternal well-being in pregnancy. It is not possible to decide between the two explanations using available data. The association of low birth weight with use of anti-hypertensives, ecclampsia/pre-ecclampsia, heart disease, placenta previa and pyrexia is explained by their effect on placental function.  The finding that women who get their ante-natal care from private clinics and private physicians have a higher risk for low birth weight is difficult to explain because these women are expected to be of a higher socio-economic status and thus to have a better pregnancy outcome. Hypotheses on the cause of this paradoxical finding are: (i) general practitioners may not provide comparable ante-natal care as is provided in government set-ups (ii) women with pregnancy complications go to general practitioners for their ante-natal care (iii) there could be unknown lifestyle factors and stresses that explain the findings. The consistent finding of differences between cases and controls in weight, blood pressure, pulse rate and fundal height on follow-up in the ante-natal period indicates that the factors of low birth weight operate throughout pregnancy; low birth weight is a result of cumulative effects and not accidental ones. The finding of association between complicated labor and risk of low birth weight confirms what is already known in the literature. Of maternal factors only 2 were of borderline significance. Marriage at ages below 20 years was associated with lower risk for low birth weight which is expected if pregnancy occurs in the late teens for example 18-19 years. Age at marriage could be a proxy for parity or maternal age. Its relationship to socio-economic status is complex and variable. The finding of lower risk for women on contraception  is a reflection of the higher socio-economic status, education and higher health-consciousness among those who plan their families. This is a preliminary study whose findings will be confirmed in a prospective follow-up study that is being planned in Kuantan. Preventive interventions can only be based on results of such a study as recommended by Chia (5) The study will have a balanced stratified design based on potential confounders found in this study. A prospective follow-up study will also collect higher quality information on confounders. The proposed study will also allow examination of interactions since risk factors are known to interact with one another (Abel 1997). The prospective study will also be able to use higher quality data to study the effect of gestation age, gravidity and parity. The study will also include weighing and examining the placenta.

 

ACKNOWLEDGEMENT

Prof Dr Md Tahir Azhar, Dean of the Kulliyah of Medicine at UIA, is gratefully acknowledged for initiating the study, directing it, and following it up until conclusion. Also acknowledged is the Director of Health, State of Pahang; Director of the Kuantan General Hospital; the head and staff of the department of Obstetrics and Gynecology, Kuantan General Hospital, for their contribution to the success of the study.

 

ADRESS FOR REPRINTS

Prof Dr Omar Hasan Kasule, Sr. Kulliyah of Medicine, International Islamic University PO Box 141 Kuantan 25710 Pahang, Darul Makmur, Malaysia e-mail: kasule@iiu.edu.my

 

SUBMITTED: 1st December 1998

 

REFERENCES

 

Yadav H: Low Birth Weight Incidence in Lundu, Sarawak. Med Journal of Malaysia 1994 49(2):164-168

 

Malaysian Very Low Birth Weight Study Group: A  National Study of Risk Factors Associated With Mortality in Very Low Birth Weight Infants in the Malaysian Neonatal Intensive Care Units. Journal of Pediatric Child Health 1997 33(1): 18-25

 

Boo NY. Outcone of very low birthweight neonates in a developing country: experience from a large Malaysian maternity hospital Singapore Med J 33:33-37, 1992

 

Boo Nem Yun: Comparison of Morbidities in Very Low Birth weight and Normal Birth weight Infants During the First Year of Life in a Developing Country. Journal of Pediatric Child Health 32(5): 439-44, 1996

 

Chia CP: Low Birth Weight Babies. Medical Journal of Malaysia 1995 50(1):120

 

Marlia Mohammed Salleh: A Study of a Sample of Low Birth Weight Babies and The Factors Which May Affect the weight of these babies. Universiti Malaya Thesis 1986/87.

 

Bailah Leigh: Social Deprivation, an Obstetric Problem?: A Study of Obstetric Outcome in Squatter Settlements in Kuala Lumpur. Thesis, Institute of Child health, London 1982.

 

Markson-LE; Turner-BJ; Houchens-R; Silverman-NS; Cosler-L; Takyi-BK : Association of maternal HIV infection with low birth weight [see comments]. J-Acquir-Immune-Defic-Syndr-Hum-Retrovirol. 1996 Nov 1; 13(3): 227-34

 

Goldenberg-RL; Cliver-SP; Mulvihill-FX; Hickey-CA; Hoffman-HJ; Klerman-LV; Johnson-MJ. Medical, psychosocial, and behavioral risk factors do not explain the increased risk for low birth weight among black women. Am-J-Obstet-Gynecol. 1996 Nov; 175(5): 1317-24

 

Bener-A; Abdulrazzaq-YM; Dawodu-A.  Sociodemographic risk factors associated with low birthweight in United Arab Emirates. J-Biosoc-Sci. 1996 Jul; 28(3): 339-46

 

McFarlane-J; Parker-B; Soeken-K. Abuse during pregnancy: associations with maternal health and infant birth weight. Nurs-Res. 1996 Jan-Feb; 45(1): 37-42

 

Spinillo-A; Capuzzo-E; Iasci-A; Nicola-S; Piazzi-G; Baltaro-F.  Sociodemographic and clinical variables modifying the smoking-related risk of low birth weight. Int-J-Gynaecol-Obstet. 1995 Oct; 51(1): 15-23

 

Liang-J; Wu-Y; Miao-L : [Analysis of factors contributing to low birth weight in Sichuan Province. Cooperating Group for Birth Defects Monitoring]. Hua-Hsi-I-Ko-Ta-Hsueh-Hsueh-Pao. 1995 Jun; 26(2): 210-4

 

Bissot-A; Villera-K; Solano-H; Bethancourt-L; Lawson-A. [Perinatal factors which affect low birth weight]

Rev-Med-Panama. 1995 Jan-May; 20(1-2): 25-32

 

Hughes-D; Simpson-L. The role of social change in preventing low birth weight. Future-Child. 1995 Spring; 5(1): 87-102

 

Chomitz-VR; Cheung-LW; Lieberman-E.  The role of lifestyle in preventing low birth weight. Future-Child. 1995 Spring; 5(1): 121-38

 

Peabody-JW; Gertler-PJ. Are clinical criteria just proxies for socioeconomic status? A study of low birth weight in Jamaica. J-Epidemiol-Community-Health. 1997 Feb; 51(1): 90-5

 

Ceron-Mireles-P; Sanchez-Carrillo-CI; Harlow-SD; Nunez-Urquiza-RM. [Conditions of maternal work and low birth weight in Mexico City]. Salud-Publica-Mex. 1997 Jan-Feb; 39(1): 2-10

 

Abel-MH. Low birth weight and interactions between traditional risk factors. J-Genet-Psychol. 1997 Dec; 158(4): 443-56

 

Shennan-AT. Factors in low birth weight [letter]. CMAJ. 1997 Dec 15; 157(12): 1737-9

 

Solla-JJ; Pereira-RA; Medina-MG; Pinto-LL; Mota-E. [Multifactorial analysis of risk factors for low birth weight in Salvador, Bahia]. Rev-Panam-Salud-Publica. 1997 Jul; 2(1): 1-6

 

Ministry of Health Annual Report 1997, Kuala Lumpur Malaysia


 

 

 

TABLE 1: MAJOR ITEMS INCLUDED IN THE STUDY QUESTIONNAIRE

 

CATEGORY

I T E M S

Socio-demographic

 

Age, Place of birth, Current residence (town, district, state), period at current address, ethnic group, religion, occupation (mother & father), monthly income (mother and father), family dwelling (type, material, ownership), type of transportation used/owned

 

Current pregnancy

Major complaints, drug ingestion, infections, traumatic injuries, stressful events, pregnancy complications, adequacy of ante-natal care (site, provider, trimester of 1st visit, total number of visits), clinical monitoring at each visit (weight, blood pressure, pulse rate, fundal height), laboratory assessment (hemoglobin, hematocrit, urine protein, and VDRL)

 

Labour &

Delivery

 

Type (spontaneous or induced), type of induction (medical or surgical), complications,  gestation age at delivery (full-term, pre-term, and small for gestation age), method of delivery (SVD or CS), outcome (healthy LB, stillbirth, neonatal mortality, neonatal morbidity), weight at birth, delivery complications, congenital malformations

Maternal factors

Age at menarche, marital status, age at first marriage, years of formal schooling, social problems (dadah, alcohol, cigarette smoking, no close or family friends), height, birth control before current pregnancy, weekly frequency of various foods before and after pregnancy, family history of non-communicable disease, clinical examination (supine BP, sitting BP, pulse, respiratory rate, heart sounds, heart murmurs, chest auscultation) 

 

Obstetric history

Live-births, abortions, still-births, termination of previous pregnancies (induction, SVD, CS), complications of previous pregnancies

 

Gynecologic history

Record of the most significant recent diagnoses

 

Medical history

Record of the most significant recent diagnoses

 

Surgical history

Record of the most significant recent diagnoses

 

 


TABLE 2: SUMMARY TABLE OF COVARIATES SIGNIFICANTLY ASSOCIATED WITH LOW BIRTH WEIGHT

 



C    A    S     E    S

C  O  N  T  R  O  L  S

ASSOC

EFFECT MEASURE

Risk Factor

YES

N

%

NO

N

%

 

TOTAL

YES

N

%

NO

N

%

 

TOTAL

 

CHI-SQ

p-value

 

OR: CRUDE

95% CI

SES: Address: Zip 25150

11

11.1

88

88.9

99

10

5.0

190

95.0

200

3.787

0.052

2.2

1.0, 5.0

 

SES: Husband Occup: clerical & General

18

18.2

81

81.8

99

11

5.5

188

94.5

298

12.051

0.001

3.2

1.7, 6.5

 

SES: Transportation: Motobike

44

52.4

40

47.6

84

63

37.7

104

62.3

167

4.909

0.027

1.4

1.0, 1.9

 

Use of anti-hypertensives

18

18.4

80

81.7

98

4

2.0

195

98.0

199

25.616

0.001

9.1

3.9, 21.6

 

Ecclampsia/pre-ecclampsia

20

20.2

79

79.8

99

10

5.5

188

95.0

198

16.685

0.001

4.0

2.1, 7.8

 

Heart disease

2

2.0

96

98.0

98

0

0

200

100

200

4.109

0.043

10.1

0.5, 209.4

 

Placenta previa

5

6.1

77

93.9

82

1

0.6

166

99.4

167

7.071

0.008

10.2

1.8, 56.5

 

Pyrexia

6

7.2

77

92.8

83

0

0

167

100

167

12.369

0.001

26.0

1.4, 456.0

 

Private clinic/physician

 

11

11.1

88

88.9

99

5

2.5

194

97.5

199

9.620

0.002

4.4

1.7, 11.3

 

Pre-term labour

 

32

38.1

52

61.9

84

4

2.4

163

97.6

167

57.978

0.001

15.9

7.8, 32.5

 

Pre-maturity pregnancy 3

 

2

9.5

19

90.5

21

0

0

72

100

72

7.008

0.008

18.6

0.9, 403.3

 

Placenta previa pregnancy 3

2

9.1

20

90.9

22

0

0

72

100

94

6.688

0.010

17.7

0.8, 283.1

 

Mole pregnancy 1

 

2

4.8

40

95.2

42

0

0

133

100

133

6.407

0.011

16.5

0.8, 350.0

 

Age at 1st marriage <20

29

29.9

68

70.1

97

83

41.7

116

58.3

199

3.868

0.049

0.6

0.4, 1.0

 

Use of contraceptives

12

14.3

72

85.7

84

41

24.7

125

75.3

166

3.620

0.057

0.6

0.3, 1.0

 

Omar Hasan Kasule, Sr December 1998