Indmedica Home | About Indmedica | Medical Jobs | Advertise On Indmedica
Search Indmedica Web
Indmedica - India's premier medical portal

Indian Journal of Community Medicine

Perinatal Mortality in Employed Women

Author(s): Bratati Banerjee, Pronab Chatterjee, Tushar Kanti Dey

Vol. 28, No. 3 (2003-07 - 2003-09)

Deptt. of Public Health Administration, All India Institute of Hygiene and Public Health, Kolkata - 700 029


Research question: Is employment during pregnancy a risk factor for perinatal mortality?

Objective: To study the occurrence of perinatal mortality in working women and compare the results with those of non-working women taken as controls and also between the experience of working women before and after joining service. Study design: Retrospective cohort study.

Setting: Jute mills in Hooghly district of West Bengal.

Participants: 100 women workers of the mills under study as study population and 100 non-working wives of male workers of the same mills as control population.

Statistical analysis: Mortality rates, Z test of proportion, Z2 test of significance.

Results: Still birth occurred more often after joining service than before (p<0.05) and more in working women in general than controls, but the latter was not statistically significant. Early neonatal death rate was more among controls than the working women and within the latter group, more before joining service than after. On controlling for gravidity and comparing stillbirth in working women after joining service, with controls, it showed higher rates in study group till 5th gravida, but only for 1st and 2nd gravida differences were significant (p<0.05). Perinatal mortality also occurred mainly in first 3-4 orders of birth after joining service.

Still birth rate in the control group was higher with birth spacing <2 years than with spacing >2 years, but in the study group it was lower in case of the former than the latter. Early neonatal death was more in pregnancies with spacing <2 years in both the groups. Conclusion: There is some risk of perinatal loss, especially still birth, existing in the working women.

Key Words: Perinatal mortality, Stillbirth, Early neonatal death, Employed women


With increasing industrialization, there has been an increase in the working population who are exposed to various health hazards. The women work force, along with the general health hazards, encounter certain work exposures which are of risk during pregnancy. Research has shown these work exposures to affect the growing fetus in several adverse ways, often resulting in fetal loss.

Heavy lifting, standing long hours, irregular working hours, shift work and extremes of heat and cold have all been reported to be associated with adverse outcomes such as spontaneous abortion, stillbirth, premature delivery, low birth weight and congenital malformations1.

Perinatal mortality (PNM) includes both late fetal deaths i.e. still births (SB) and early neonatal deaths (END). These two types of deaths are combined because the factors responsible for both are often similar, being those operating before and around the time of birth2. Two important factors responsible for PNM are birth weight and period of gestation, with which it was found to be inversely proportional3.

Material and Methods:

A retrospective cohort study was conducted, in jute mills of Hooghly district of West Bengal. The study was carried out over a period of six months, but the total reproductive experience of subjects, till the date of interview, was considered.

Women in the reproductive age group were taken as the study population. Thus, women aged 50 years and above were excluded from the study to avoid recall bias. Widows, who had been employed in the mills on compassionate ground after the demise of their husbands, were not included in the study as they were not likely to become pregnant again, during their period of work in the mill. Four jute mills, located in the southern part of the district, were selected for the study and all the women workers, eligible to be included in the study group who could be contacted, were interviewed. The number of study subjects thus selected came to 100. As fresh recruitment of women workers in heavy work has been stopped now a days, all the women were above 35 years of age. Thus the study group comprised of 100 women within the age of 35-49 years.

The control cohort also comprised of 100 women, who were the non-working wives of the men working in the same jute mills, and were of the same age group, from similar socio-economic status, living under similar conditions.

The study was conducted by face-to-face interview of both the groups, according to a pre-tested, semi-structured schedule, which was designed to cover the general data and detailed occupational and obstetric histories. Office records, insurance cards and available medical records were consulted for verification of information.


Table I: Types of perinatal mortality in the total population.

Study population Control group Study group Risk Ratio
Before employment After employment Risk ratio
Total pregnancies 265 270   566 535  
Still births (SB) 2 8 3.93 8 10 1.32
SB rate (%) 0.76 2.96*   1.41 1.87**  
Early neonatal death (END) 8 3 0.36 24 11 0.49
END rate (%) 3.02 1.11   4.24 2.06  
Perinatal mortality (PNM) 10 11 1.08 32 21 0.69
PNM rate (%) 3.77 4.07   5.65 3.93  

*Z=1.88, p < 0.05; **Z=0.61, p > 0.05

It may be observed from Table I that SB occurred often in employed women after their joining service than before and this was statistically significant (p<0.05). Though SB rate was higher in the working group than the control group, the difference was not significant on testing (p>0.05). Yet, in this case also, risk ratio was raised much more (RR=1.32). Early neonatal mortality showed just the reverse result. Rate was more in the control group than the study group, and within the study group, rate was more before joining service than after.

Table II: Stillbirths in the study group after joining service in comparison to the control group according to gravidity.

Gravida Study group Control group Odds ratio
Total pregnancy Stillbirths Stillbirth rate Total pregnancy Stillbirths Stillbirth rate
>lst 74 3 4.05 566 8 1.41 2.95*
>2nd 113 4 3.54 466 6 1.29 2.81
>3rd 128 2 1.56 369 5 1.35 1.16
>4th 105 3 2.86 279 5 1.79 1.61
>5th 101 3 2.97 199 4 2.01 1.49
>6th 79 1 1.27 132 2 1.52 0.83
>7th 52 1 1.92 81 2 2.47 0.77
>8th 38 - - 46 1 2.17 -

*p < 0.05

Table II shows the comparison of the reproductive performance of the working women after joining service directly with that of the non-working women. For the study group, all the pregnancies occurring following each order of gravidity after joining service, were considered for determining the SB rate, thus altering the housewife status of these subjects, before they joined service. For the control group, the SB rate among pregnancies following every order of birth, were calculated. The two groups were then compared for SB rate following each gravida, thus eliminating the effect of gravidity. It was seen that odds ratio (OR) was raised till the 5th gravida. Thereafter, it decreased to below 1, thus indicating higher SB rates in the control group for higher orders of births. On testing, the differences were found to be just significant (p<0.05) only for 1st and 2nd gravida.

Table III: Perinatal mortality in the working women according to order of birth after joining service.

Order of
birth after
Total pregnancy Still births Early neonatal deaths Total perinatal mortality
No. (%) No. (%) No. (%)
1st 86 4 (4.65) 1 (1.16) 5 (5.81)
2nd 64 2 (3.13) 1 (1.56) 3 (4.69)
3rd 49 1 (2.04) 0 (-) 1 (2.04)
4th 33 0 (-) 1 (3.03) 1 (3.03)
5th 16 0 (-) 0 (-) 0 (-)
6th 9 0 (-) 0 (-) 0 (-)
7th 7 1 (14.29) 0 (-) 1 (14.29)
>8th 6 0 (-) 0 (-) 0 (-)
Total 270 8 (2.96) 3 (1.11) 11 (4.07)

Perinatal mortality was seen to have occurred only in the first 3-4 orders of birth after joining service and this too showed a declining trend with increasing orders of birth (Table III).

Table IV: Perinatal mortality rate in the study and control group according to birth spacing

Birth spacing Stillbirths Early neonatal deaths Perinatal mortality
Study Control Study Control
Study Control
<2 years 2.38 1.73 1.59 5.19 3.97 6.93
>2 years 3.15 0.85 0.78 2.98 3.93 3.83

SB rate in the control group was observed to be higher in pregnancies with birth spacing <2 years than with spacing >2 years, whereas, in the study group, it was lower in case of the former than the latter. END rate, however, was found to be higher in case of spacing <2 years in both study and control groups.


Stillbirth was observed to occur more in the working women than the control group and within the study group more after joining service than before. But total PNM rate in the employed women, though higher after joining service than before, was in general, lower than the control group.

McDonald et al, in their study on 6 occupational groups, observed that risk ratios for SB without defect were increased considerably. 0/E ratio for SB was 1.13 in the manufacturing sector, and within this sector, 1.90 for textile industry4. This study was further followed up in 1988 which again supported the previous findings. 0/E ratios for SB (2.22) was significantly higher in textile industry (p<0.01)5. Najman et al reported the perinatal death by mothers' employment status. He observed PNM rate in employed mothers to be 10.5/1000 pregnancies, while that in housewives was 9.4, but this difference was not significant6. Stewart reported that working mothers had higher rates of perinatal deaths and this rate was higher in women who worked after the 28th week of gestation7.

However, other researchers have failed to find any difference in perinatal death rates of working and non working women, while some have found even contradictory results8,9. A comparative study between housewives and working women by Marbury et al showed that PNM rate was same in both the groups8. In yet another study the PNM rate was higher for unemployed than for employed mothers (p<0.05) and also higher for employed mothers who stopped work before 30 weeks of pregnancy, than for those who continued to work after 30 weeks (p<0.05). Yet within the employed group, mothers who were involved in non-sedentary work experienced higher PNM rates than the sedentary worker groups, though this difference was not statistically significant. This paradoxical finding was explained by the fact that the authors observed that unemployed women were more likely to be at the extremes of maternal age, had a history of medical problems and had attended AN care less frequently9.

Physical effort during pregnancy is a known risk factor. McDonald et al observed that O/E ratios for SB were raised with physical effort (1.87, p<0.01) and with standing (1.46, p<0.05)5. Senturia's results revealed that certain work factors directly correlated with miscarriage and/or perinatal death. The significant factors included: fewer household helpers, standing, working in hot environment, commuting, walking, carrying and lifting heavyweight on the job10. Heavy weight lifting during pregnancy had a detrimental effect on the fetus, especially late in pregnancy11.

Early neonatal death showed results reverse of SB, END rate was higher in non-working than working women and in the latter group, higher before their joining service than after. This may be because working women, by supplementing their family income, raise the economic status of their families and Ramji, on comparing various studies in India, has shown that neonatal mortality is lower in higher income households12.

An American study, using data obtained from National Survey in 1980, showed that a history of non live birth outcomes was consistently more common among working women. Of this, history of stillbirth was more common among working women, principally for women of gravidities 2 and 313. A significant association between gravidity and stillbirth in the working women was also observed in the present study. By eliminating the non working period in the lives of the working women and also by controlling for gravidity, it was seen that SB rate was more in the study group till the 5th gravida, but on testing, only the difference for 1st and 2nd gravida proved to be significant (p<0.05). Within the working group, PNM was found to occur only in the first 3-4 orders of birth after joining service and a declining trend was observed with increasing orders of birth. This may indicate that the risk is highest immediately after joining service, when the woman is exposed to unaccustomed fatigue. The risk then gradually falls as the woman adjusts herself to the work and gets habituated with the workload.

Shorter birth intervals of less than 2 years is a known risk factor for pregnancy2. The effect of less birth spacing on PNM was studied and it was observed that PNM rate in case of less birth spacing was almost double that of pregnancies with spacing more than 2 years. However, in the study group, the rates were almost the same in both the groups. The failure to detect any difference in perinatal outcome in the employed group may indicate that the pregnancy outcome in these women was not influenced by birth spacing, probably because some other factor i.e. their work exposure, might have been more important in determining the outcome.

The results of present study give an indication of some risk of perinatal loss, especially stillbirth, existing in the working women. Reproductive hazards are of such grave concern that such an issue should be resolved by conducting prospective studies, which will establish cause and effect relationship, so that measures may be taken to reduce the risk and ensure healthy infant and mother.


  1. l. Hunter's Diseases of Occupations: 8th Ed.(ED) PAB Raffle, PH Adams, PJ Baxter and WR Lee. Edward Arnold Publishers, London 1994: 734.
  2. K. Park. Park's Textbook of Preventive and Social Medicine: 16th Ed, M/s. Banarsidas Bhanot Publishers, Jabalpur, India 2000: 384.
  3. Agarwal VK, Gupta SC, Roy Chowdhury S et al. Some observations on perinatal mortality. Indian Paediatrics 1982;19(3): 233-8.
  4. McDonald AD, McDonald JC, Armstrong B et al. Occupation and pregnancy outcome. Br J Int Med 1987; 44: 521-6.
  5. McDonald AD, McDonald JC, Armstrong B et al. Fetal death and work in pregnancy. Br J Ind Med 1988; 45:145-57.
  6. Najman JM, Morrison J, Williams GM, Andersen MJ, Keeping JD. The employment of mothers and the outcomes of their pregnancies: An Australian Study. Public Health 1989;103: 189-98.
  7. Steward A. A note on the obstetric effects of work during pregnancy. Br J Prev Soc Med 1955; 9:159-61.
  8. Marburry MC, Linn S, Monson RR et al. Work and pregnancy. J Occup Med 1984; 26(6): 415-21.
  9. Murphy JF, Dauncey M, Newcombe R, Garcia J, Elbourne D. Employment in pregnancy: prevalence, maternal characteristics, perinatal outcome. Lancet May 26, 1984;1(8387): 1163-6.
  10. Senturia KD. A woman's work is never done: Women's work and pregnancy outcome in Albania. Med Anthropol Q 1997;11(3): 375-95.
  11. Banerjee B, Dey TK, Chatterjee P. Estimation of risk of pregnancy wastage due to lifting of heavy weight during pregnancy. Indian J Occup Environ Med 2002; 6(1): 13-5.
  12. Ramji S. Socio-economic and environmental determinants of perinatal and neonatal mortality in India. Indian Paediatrics 1989; 26(11): 1100-5.
  13. Savitz DA, Whelan EA, Rowland AS, Kleckner RC. Maternal employment and reproductive risk factors. Am J Epidemiol 1990; 132(5): 933-45.
Access free medical resources from Wiley-Blackwell now!

About Indmedica - Conditions of Usage - Advertise On Indmedica - Contact Us

Copyright © 2005 Indmedica