The influx of expectant mothers from Mainland China leading to overwhelming of the local obstetric and neonatal services has been a hot topic of discussion in the media in the past few years. In 2001, the Hong Kong Court of Final Appeal delivered a unanimous opinion by which Chong Fung-yuen, a Chinese baby born while his Mainland Chinese parents were in Hong Kong on two-way permits, was granted residency in Hong Kong. In addition, in 2003, the Hong Kong SAR Government introduced the Individual Visit Scheme which allowed travellers from Mainland China to visit Hong Kong and Macao on an individual basis. Since the introduction of these policies, there has been a dramatic increase in the number of ‘traveller’ mothers delivering in Hong Kong. This ‘birth tourism’—a travel to a country that practises birthright citizenship or permanent residency in order to give birth there, so that the child will be a citizen of the destination country—has significantly influenced the standard obstetric practice in Hong Kong, resulting in adverse pregnancy outcomes. We have described this phenomenon as social obstetrics.1 2 3

According to the Hospital Authority (HA) pay code, there are seven categories of non-eligible person (NEP). The categories of NE-2 (People’s Republic of China passport or two-way permit holder ‘雙非’) and NE-3 (NEP whose spouse is a Hong Kong Identity Card [HKID] holder ‘單非’) contribute to the majority of NEP deliveries in public hospitals.

In 2005, the HA launched an obstetric package to limit the number of non-local women delivering in public hospitals. The charge was HK$20 000 for 3 days and 2 nights of hospital stay including delivery. However, this policy did not discourage ‘traveller’ mothers from delivering in public hospitals.4

In February 2007, the HA launched a new obstetric package for non-local expecting mothers. This package charged almost double (HK$39 000) for the hospitalisation for 3 days and 2 nights. Those who have not booked were additionally charged.

Unfortunately, the growing number of NEP deliveries in HA hospitals outweighed the capacity of public obstetric and neonatal services. In 2010, there were about 88 000 deliveries in the territory, of which 50% were by mainland mothers (Fig 1a). The total capacity of neonatal intensive care units (NICUs; about 100 beds) in Hong Kong can only support an annual delivery rate of 75 000.5 This resulted in the formation of Hong Kong Obstetric Service Concern Group in March 2011—to urge the Hong Kong SAR Government to take action in preventing the collapse of public obstetric and neonatal services. The first remedial measure was to stop accepting new antenatal booking in HA hospitals from 8 April 2011 till the end of the year. One year later, on 26 April 2012, HA announced that there was no booking quota for non-local expecting mothers as public service was prioritised for the Hong Kong citizens to meet the surge of childbirth in the Chinese year of ‘Dragon’. The non-booked deliveries would be charged HK$90 000 for the 3-days-2-nights package. Lastly, the Government prohibited antenatal booking of non-local mothers in either public (for NE-2 and NE-3 categories) or private (for NE-2 category) sectors from 1 January 2013 onwards (Table 16 7).

However, if a pregnant woman, regardless of her identity card status, attended the accident and emergency department (AED) of a public hospital, the doctor-on-duty would assess her condition and offer admission to the obstetric unit if medically indicated. The admission rate via AED through the years varied with the implementation of obstetric package and government policy (Fig 1b). In 2005, when the first obstetric package was launched, the admission through AED for delivery was high. The second package in 2007 encouraged antenatal booking, and, thus, the AED admission rate dropped thereafter. Since April 2011, HA stopped all antenatal bookings for NEP, as a result of which the total number of NEP deliveries decreased drastically; however, the proportion of AED admissions increased.

In Kwong Wah Hospital, antenatal booking for non-local mothers had been stopped since April 2011. The NEP deliveries in our unit were mainly through AED admission or transfer from another HA or private hospital.

This retrospective study reviewed the pregnancy conditions and outcomes of a cohort of non-booked, non-local women admitted via AED of Kwong Wah Hospital over a 1-year period.

This study evaluated the demographics, peripartum events, and pregnancy outcomes of the non-local pregnant women (NE-2 and NE-3 categories) who were admitted through AED and who delivered in Kwong Wah Hospital from 1 April 2011 to 31 March 2012. This was the 1-year period after HA’s announcement (on 8 April 2011) of stopping antenatal booking for non-local women. The birth registry record of Kwong Wah Hospital was reviewed. Women who delivered in the captioned period with no HKID number were identified. Clinical records of the subjects were retrieved from the central record unit. Only women in NE-2 and NE-3 categories were recruited.

Clinical notes and electronic patient records of the subjects were reviewed. The pregnancy conditions studied included the presenting symptoms, antenatal complications, gestation at delivery, mode of delivery, intrapartum and postpartum complications, birth weight of babies, Apgar score, need for neonatal resuscitation, admission to NICU or special baby care unit, neonatal morbidities, congenital abnormalities, etc. Maternal and neonatal outcomes were further analysed according to their booking status before admission. The annual statistics of Kwong Wah Hospital 2011 were used as reference.

Skewed continuous variables and nearly normally distributed variables were presented as medians (interquartile ranges) and means (± standard deviations [SDs]), respectively. Categorical data were presented as counts and percentages. Mann-Whitney U test and independent sample t test were used for comparison of medians and means, respectively. Pearson Chi squared test or Fisher’s exact test were used for comparisons of frequencies, where appropriate. All analyses were performed with the Statistical Package for the Social Sciences (Windows version 16.0; SPSS Inc, Chicago [IL], US). A P value of less than 0.05 was regarded as statistically significant.

Ethics approval for this study was granted by the Kowloon West Cluster Clinical Research Ethics Committee.

A total of 219 maternities with delivery were identified during the study period. There were 221 live births (three pairs of twins) and one stillbirth. Two (0.9%) pregnancies had been achieved by assisted reproduction. The mean (± SD) age of women was 29.9 ± 5.6 years. Of the 219 women, 138 (63.0%) were multiparous, 28 (12.8%) of them had had one previous caesarean delivery, and one (0.5%) had had two previous Caesarean sections. Overall, 139 (63.5%) women were of NE-2 category and 53 (24.2%) were of NE-3 category; the remaining 27 (12.3%) did not provide information about their partners. A total of 138 (63.0%) women had no booking in Hong Kong; 61 (27.9%) women had antenatal booking in other HA hospitals; and 20 (9.1%) women were booked in private hospitals but were referred or chose to deliver in HA hospitals.

The reasons of admission were as follows: show or with irregular uterine contraction (n=98, 44.7%), suspected rupture of membranes (n=52, 23.7%), active phase of labour (n=40, 18.3%), and antenatal complications (n=21, 9.6%; these included 10 cases of antepartum haemorrhage, five cases of preterm prelabour rupture of membranes, three cases of concerns on fetal wellbeing, two cases of maternal pre-eclampsia, and one case of threatened preterm labour). Five women admitted for postdate pregnancy requested for delivery. Two pregnancies were delivered in an ambulance and one on arrival to AED. One pregnancy was a stillbirth diagnosed after admission.

Routine antenatal blood tests (complete blood picture, blood group and Rhesus factor, immune status for hepatitis, syphilis, rubella, and human immunodeficiency virus) were performed in 147 (67.1%) women. For the rest of the women, results of blood tests performed in another HA or private hospital were available via electronic or hard copies. Ultrasound assessment was performed for 126 (57.5%) women before delivery.

Of the 219 pregnancies, 23 (10.5%) were delivered before 37 weeks of gestation; two (0.9%) pregnancies were delivered after 42 weeks of gestation. A total of 141 (64.4%) women had spontaneous onset of labour; 32 (14.6%) needed induction of labour, and 22 (10.0%) needed augmentation of labour.

The majority of women (n=182; 83.1%) had normal vaginal deliveries. Three (1.4%) pregnancies required instrumental assistance. Caesarean section was performed in 13 (5.9%) pregnancies after labour and 21 (9.6%) without labour. The success rate of trial of vaginal delivery after one previous Caesarean section was 50%. There was no uterine scar rupture in any case. Primary postpartum haemorrhage occurred in 13 (5.9%) pregnancies. Seven (3.2%) women required blood transfusion. The mean length of postnatal hospital stay was 2.0 ± 0.4 days.

Peripartum maternal complications were divided into mild and significant. Mild complications included seven cases of gestational hypertension, two cases of mild pre-eclampsia without magnesium sulphate treatment, three cases of gestational diabetes on insulin treatment, three cases of moderate thrombocytopenia (platelet count 50-100 x 10⁹ /L), five cases of retained placenta requiring surgical exploration, five cases of postpartum haemorrhage managed by medical therapy, three cases of post-delivery urinary retention, and five cases of postpartum fever. Significant complications included six cases of severe pre-eclampsia requiring magnesium sulphate treatment, two cases of placenta abruptio, two cases of major placenta praevia type IV, one case of massive primary postpartum haemorrhage requiring surgical intervention, and two cases of severe thrombocytopenia (platelet count <50 x 10⁹ /L).

During the study period, there were 121 (54.8%) male and 100 (45.2%) female live births. The mean birth weight was 3.3 ± 0.5 kg. There were 19 (8.7%) babies with low birth weight (<2.5 kg); 13 (5.9%) were macrosomic (>4.0 kg). Two babies required neonatal resuscitation. The admission rates to the NICU and special care baby unit (SCBU) were 3.7% and 43.8%, respectively. Overall, 15 (6.8%) babies had minor congenital abnormalities. Three (1.4%) had major abnormalities, including one ventricular septal defect, one atrial septal defect, and one bilateral congenital cataract. Apart from congenital problems, 51 babies had neonatal jaundice requiring phototherapy, 22 had respiratory complications, 22 had infection episodes, five had electrolyte disturbance, three had birth trauma, three had congenital hypothyroidism, three had hypoglycaemia, one had hypothermia, one had polycythaemia, one had anaemia requiring blood transfusion, one had neonatal autoimmune thrombocytopenia requiring intravenous immunoglobulin treatment, and one had neurological complications. The composite neonatal morbidity rate was 39.8%.

The pregnancy outcomes of the study cohort were compared with the annual statistics (2011) of Kwong Wah Hospital, as shown in Table 2. Non-local mothers were of higher parity; more likely to have hypertensive disease (including pre-eclamptic toxaemia), preterm delivery (<37 weeks), babies requiring admission to SCBU, and macrosomic babies (>4.0 kg). The rate of induction of labour and caesarean section was lower in this group.

We also analysed the maternal and neonatal outcomes based on their antenatal booking before admission (ie no booking versus booking in other HA or private hospitals). We found that there was no significant difference in maternal and neonatal outcomes between the two groups. The results are shown in Table 3.

Standard obstetric practice is influenced by social behaviour such as ‘birth tourism’ resulting in adverse pregnancy outcomes; we have described this phenomenon as social obstetrics.1 2 3 Some women came because they wanted to evade the ‘one-child’ policy of Mainland China. This was reflected in our study which showed that 63% of the NEP mothers were multiparous versus 45% from the hospital annual statistics. The higher proportion of multiparity also explained the lower rate of labour induction and caesarean delivery in the study group. On the other hand, the significantly higher rates of preterm delivery, hypertensive disease, macrosomic babies, and SCBU admission suggest that the NEP mothers belonged to a high-risk group.

In this study cohort, 63.5% of women belonged to the NE-2 category. Their travelling permit only allowed a short period of stay. In principle, there could be shared care between Hong Kong and Mainland China; in reality, this form of shared care is often suboptimal because of the differences in clinical practice and culture between the two places. Some mothers could not make antenatal booking in Mainland China under the ‘one-child’ policy. Serious conditions may be detected for the first time during an emergency admission.8 This largely endangers the health of mothers and babies. We have chosen six typical cases for illustrating this issue (Table 49 10 11 12).

Over the years, two local studies have been published on the pregnancy outcomes of non-local expectant mothers delivering in public hospitals in Hong Kong.6 7 Yuk and Wong6 from Princess Margaret Hospital conducted a study between 2004 and 2006 when the HA launched the first obstetric package to the non-local women in 2005. During that period, around 35% of deliveries in Princess Margaret Hospital were attributed to non-local Chinese women. The proportion increased significantly from 27% in 2004 to 43% in 2006. Compared with local Chinese women, the NEPs were younger, of lower parity, and had fewer pre-existing medical problems. However, they had higher chances of unplanned vaginal breech deliveries, severe hypertensive disease in pregnancy, pre-eclampsia, delivering before arrival to hospital, and giving birth post-term (≥42 weeks). Neonatal complications including preterm birth, stillbirth, and neonatal death were also more frequent among the NEP women. In fact, the first obstetric package was proposed mainly to charge the NEPs for delivery service expenses; it did not cover the antenatal service. This resulted in many of them coming to the hospital only for giving birth. Many of them came at the ‘last-minute’ to reduce the length of hospital stay due to financial concerns. This created a heavy burden on the public obstetric services and increased the risk of adverse pregnancy outcomes.

The second obstetric package in 2007 encouraged the NEP mothers to receive proper antenatal checkup. Lam7 from Tuen Mun Hospital conducted a study from 2006 to 2008 investigating the impact of the package on public obstetric services and pregnancy outcomes. It was observed that the number of NEP deliveries decreased from 1868 to 1398 per year. The number of non-booked admissions through AED reduced. The rate of post-term pregnancies dropped from 3.2% to 1.8%. The reason for fewer deliveries was a shift of patients to private obstetric services after setting the quota and raising the cost. Nevertheless, this obstetric package did not improve the admission behaviour and pregnancy outcomes.

Thanks to our HA and the Hong Kong SAR Government’s policy of stopping NEP bookings altogether in HA Obstetric Units, the number of NEP deliveries during our study period (2011-2012) was significantly reduced and limited to non-booked cases admitted through AED (Fig 2). We observed that reduction in ‘quantity’ did not improve the ‘quality’ of care in this group of women. The admission pattern and pregnancy outcomes remained similar to those in previous local studies. We also observed that, although some women had prior ‘booking’ in other HA or private hospitals, their pregnancy outcomes were no better than the ‘no booking’ group (Table 3). One possible reason for this could be that their travelling permit hindered them from receiving the ‘standard’ antenatal care. It was difficult to measure the quality of the obstetric care received by the ‘booked group’ because of its heterogeneity. We have used case examples to illustrate how the common obstetric conditions could be ‘near-miss’ conditions or standard obstetric service could be compromised under this social obstetrics phenomenon. We foresee that the third obstetrics package introduced in 2012 is unlikely to make a significant improvement in pregnancy outcomes unless the NEP women attend a structured antenatal care like the local mothers do.

In our study, we compared the pregnancy outcomes of our NEP cohort admitted through A&E (n=219) with the general pregnant population from our annual statistics (n=5862). This might introduce a pre-selection bias. Our NEP cohort was also limited by its relatively small number. In the study by Yuk and Wong,6 the pregnancy outcomes of the NEP cohort (n=4657) were compared with those of the eligible-person cohort (n=8655) from 2004 to 2006. In the study by Lam,7 the pregnancy outcomes of two NEP cohorts (n=1868 in 2006/2007 vs n=1398 in 2007/2008) were compared.

Non-local expectant mothers delivering babies in Hong Kong has become a classic social obstetrics phenomenon. There is nothing wrong with these mothers who would like to have their children to be born in Hong Kong and become permanent residents of Hong Kong. Not long ago, Hong Kong mothers wanted to give birth in the US or Canada so that their children could become citizens of those countries. The problem in Hong Kong is the large volume of pregnancies which has exceeded our obstetric and neonatal capacities, thus affecting the health care of our local pregnant mothers and neonates. Although our Government now prohibits NEP bookings in both public (for NE-2 and NE-3 categories) and private (for NE-2 category) hospitals, non-local expectant mothers continue to admit themselves through AED for deliveries. Health care professionals should continue to be prepared for managing these potential near-miss clinical situations arising from this social obstetrics phenomenon. We hope this paper serves as one of the historical records in literature for this social obstetrics phenomenon in the recent obstetric history of Hong Kong.

1. Leung WC. Social obstetrics—non-local expectant mothers delivering babies in Hong Kong. The Hong Kong Medical Diary 2009;14:13-4.

2. Leung WC, Lau WL. Cross-border families: transgression and dialogue [in Chinese]. Hong Kong: Red Publishing; 2008: 55-9.

3. Leung WC. Social obstetrics (cross-border pregnant women) [in Chinese]. Chiu MC, editor. Children, medicine, law: a comparative study of Greater China. Hong Kong: Roundtable Publishing; 2012: 40-8.

4. Au Yeung SK. Impact of non-eligible person deliveries in obstetric service in Hong Kong. Hong Kong J Gynaecol Obstet Midwifery 2006;6:41-4.

5. Leung TY, Lao T. Influx of mainland expectant mothers: a blessing or a curse? Hong Kong J Gynaecol Obstet Midwifery 2009;11:9-10.

6. Yuk JY, Wong S. Obstetrical outcomes among non-local Chinese pregnant women in Hong Kong. Hong Kong J Gynaecol Obstet Midwifery 2009;9:9-15.

7. Lam KD. Is the new obstetrics package for non-local pregnant women making a change? Hong Kong J Gynaecol Obstet Midwifery 2010;10:62-8.

8. Kwan WY, So CH, Chan WP, Leung WC, Chow KM. Re-emergence of late presentations of fetal haemoglobin Bart's disease in Hong Kong. Hong Kong Med J 2011;17:434-40.

9. Lo TK, Yung WK, Lau WL, Law B, Lau S, Leung WC. Planned conservative management of placenta accreta—experience of a regional general hospital. J Matern Fetal Neonatal Med 2014;27:291-6. CrossRef

10. Yung C, Liu K, Lau WL, Lam H, Leung WC, Chin R. Two cases of postmaturity-related perinatal mortality in non-local expectant mothers. Hong Kong Med J 2007;13:231-3.

11. Yung WK, Liu AL, Lai SF, et al. A specialised twin pregnancy clinic in a public hospital. Hong Kong J Gynaecol Obstet Midwifery 2012;12:21-32.

12. Liu AL, Yung WK, Lai SF, et al. Factors influencing the mode of delivery and associated pregnancy outcomes for twins: a retrospective cohort study in a public hospital. Hong Kong Med J 2012;18:99-107.