International Journal for Quality in Health Care, 2018, 30(6), 466–471

doi: 10.1093/intqhc/mzy041
Advance Access Publication Date: 24 March 2018
Article

Article

Evaluating the quality of antenatal care and

pregnancy outcomes using content and
utilization assessment
PING LING YEOH1,2, KLAUS HORNETZ3, NOR IZZAH AHMAD SHAUKI4,
and MAZNAH DAHLUI1
1
Julius Centre, Department of Social and Preventive Medicine, Faculty of Medicine, University of Malaya, 50603
Kuala Lumpur, Malaysia, 2Mediconsult Sdn. Bhd., A13/1/1 Jalan Ampang Utama 2/2, One Ampang Business
Avenue, 68000 Ampang, Selangor, Malaysia, 3Consultant for Health Systems Development, Lenaustrasse 15, 04157
Leipzig, Germany, and 4Institute for Health Management, Ministry of Health Malaysia, Jalan Bangsar, 59000 Kuala
Lumpur, Malaysia
Address reprint requests to: Ping Ling Yeoh, Mediconsult Sdn. Bhd., A13/1/1 Jalan Ampang Utama 2/2, One Ampang
Business Avenue, 68000 Ampang, Selangor Malaysia. Tel: +603 4253 3200; Fax: +603 4253 3199; E-mail: pingling@
mediconsult.com.my
Editorial Decision 1 March 2018; Accepted 8 March 2018

Abstract
Objective: To assess the adequacy of antenatal care (ANC) and its association with pregnancy out-
comes using an approach that includes adequacy of both utilization and content.
Design: Retrospective cohort study.
Setting and Participants: Women attending ANC at public-funded primary health clinics where
data were extracted from individual records.
Methods: Adequacy of utilization assessment was based on the concept of Adequacy of Prenatal
Care Utilization index; adequacy of content assessed the recommended routine care received by
the women according to local guidelines. Association between adequacy and pregnancy out-
comes was examined using binary logistic regression.
Main Outcome Measures: Pregnancy outcomes included preterm birth and low birth weight.
Results: Sixty-three percent of women showed higher than recommended ANC utilization; 52%
had <80% of recommended routine care content. Although not statistically significant, the odds of
preterm birth was lower among women with adequate level of utilization compared with inad-
equate (adjusted odds ratios (aOR) = 2.34, 95% confidence interval (CI) 0.45–12.16) and intensive
levels (aOR = 3.27, 95% CI 0.73–14.60). Regarding adequacy of content, women who received inad-
equate level of care content were associated with higher prevalence of preterm birth (aOR = 3.69,
95% CI 1.60–8.55).
Conclusion: The study shows inadequate content is associated with higher prevalence of preterm
birth and suggests that inadequate utilization increases the risk of preterm birth. It demonstrates
the relevance of using both utilization and content assessment in evaluating quality of ANC.
Further studies are encouraged to review the methods used.

Key words: antenatal care, utilization, content, quality of care, pregnancy outcomes

© The Author(s) 2018. Published by Oxford University Press in association with the International Society for Quality in Health Care. All rights reserved.
For permissions, please e-mail: journals.permissions@oup.com 466

Downloaded from https://academic.oup.com/intqhc/article-abstract/30/6/466/4953352

by guest
on 02 August 2018
 Evaluating antenatal care • Patient outcomes 467

Introduction excellent [11] and the number of ANC visits have increased higher
than recommended [12, 13]. Recent progress in pregnancy out-
Assessing quality of antenatal care (ANC) requires information on ser-
comes, however, has not improved accordingly; for over a decade,
vices utilization and indicators of content or quality of ANC [1–5].
maternal mortality has stagnated [14], and the prevalence of low
However, the most commonly used adequacy of prenatal care utiliza-
birth weight (LBW; compared with that of neighbouring countries)
tion (APNCU) index assesses only adequacy of care initiation and
[15], and stillbirth [16] have increased. These highlight the limita-
adequacy of number of visits [6]. The index assesses neither content of
tions of present coverage indicators and the need to initiate assessing
care provided nor risk conditions of women [3, 6].
ANC beyond these indicators, since they are not linked to outcomes.
Delivery of ANC follows standards and guidelines based on bio-
psycho-social needs of women [7, 8]. Definition of quality maternity care
[8] and risk-oriented ANC recommended by WHO [7] imply: (i) the Methods
need for rational use of ANC according to needs and, (ii) provision of
care to contribute to intended outcomes. These demand that all women Study design and population
be given a minimum set of care regardless of risk level and that additional This was a retrospective cohort study including data of women who
care are to be given based on risk condition identified. The approach had attended public-funded health clinics [17, 18]. The women and
commands that evaluation of ANC needs to assess the recommended 6 out of 58 health clinics in Selangor state were selected by a multi-
routine care received by the women and the risk condition of the women. stage sampling.
Few studies have attempted to measure utilization and content of Patient records were anonymised and de-identified prior to ana-
ANC using tools that integrated both aspects into one composite measure lysis. The study was approved by the Research and Ethics Committee
[4, 9]; however, these tools have limitations in this regard, in certain set- of University Malaya Malaysia, and the Medical Research and Ethics
tings. For example, one of these tools was designed for places with low Committee of the Ministry of Health Malaysia.
ANC attendances, for which health education was named as one ANC
element without specifying the topics of health education that should be
Definition of variables: evaluating quality of antenatal
provided [9]. Therefore, application of this tool in high ANC attendance
care
settings is challenging, while comprehensive analysis of quality of care is
Adequacy of utilization
impossible. Also, for another tool which included three interventions
This was based on the concept of the APNCU index [6]. The index
(ultrasound, blood pressure and blood screening), the authors acknowl-
was modified in this present study by raising the cut-off points of
edged the need to incorporate additional ANC interventions [4]; however,
the observed-to-expected visits ratio, to reflect the lower recom-
this is impossible owing to the tool’s design. Moreover, the tool only
mended visits of the local guidelines [18]. Figure 1 illustrates the
assessed adequacy of care initiation but not adequacy of ANC visits.
modified visits ratio cut-off points: adequate-plus (≥130% of
Consequently, in settings with increasing number of ANC visits per
expected visits—intensive utilization with ≥30% higher visits than
pregnancy, rational use of care analysis is limited. As such, the index
recommended), adequate (90–129%), intermediate (60–89%) and
derived from such a tool may not be considered as more significant
inadequate (<59%). During analysis, utilization adequacy was cate-
than others [4], since it still requires the utilization of other indices,
gorized into three categories: adequate-plus, adequate and inad-
such as the APNCU to compensate for its limitations.
equate (intermediate was grouped with inadequate). Regarding
By analysing the complex ANC process over time, this study
initiation of care, the modified index maintained the four categorical
aimed to assess the adequacy of ANC and its association with preg-
timing of the original APNCU index. Women who had the first visit
nancy outcomes, using an approach that included both adequacy of
after 17 weeks gestation were categorized as having received inad-
utilization and adequacy of content.
equate care (Fig. 1).

Background of study setting Adequacy of content

Malaysia has recorded remarkable achievement in maternal-child- Adequacy of content assessed the level of recommended routine care
health over the past decades [10]. Relevant tracers continue to be received by the women according to the MOH guidelines [13]. The

≥27 weeks
(weeks antenatal
care started)

18–26 weeks
Initiation

9–17 weeks

≤8 weeks

< 59% 60–89% 90–129% ≥130%

Observed-to-expected visits ratio
Summary Index:
Adequate-plus
Adequate
Intermediate
Inadequate

Figure 1 Modified APNCU Index adjusted for Malaysian Recommended Schedule. Source: Adapted from APNCU index [6].

Downloaded from https://academic.oup.com/intqhc/article-abstract/30/6/466/4953352

by guest
on 02 August 2018
 468 Yeoh et al.

routine care consisted of four components—physical examination,
health screening, case management and health education
(Supplementary Table S1). Weighting factor, which was consulted and
explained in another paper [17], was applied. Content adequacy was
categorized as inadequate (<80%) or adequate (≥80%).
Outcome measures
The outcome measures used in this paper was pregnancy outcomes
which included preterm birth (<37 gestational weeks at birth) and
LBW (<2500 g at birth).
Adequacy of utilization and content
Sixty-three percent had ‘adequate-plus’ utilization (Table 2), indicat-
ing a high proportion of women had higher number of ANC visits
than the recommended schedule.
Fifty-two percent of women had <80%, or inadequate level, of
recommended routine ANC content (Table 2).
There was a correlation between utilization level and mean con-
tent score (Table 3, P<0.001). The difference between women in
‘inadequate’ and ‘adequate’ utilization category was not statistically
significant (P = 0.239).

Statistical analysis Association between adequacy of utilization or content

The association between adequacy of utilization or content and and pregnancy outcomes
pregnancy outcomes was examined by binary logistic regression Adequacy of utilization and pregnancy outcomes
using a full model multivariate regression. The model was adjusted The odds for preterm birth in the adequate-plus category was 3.27
for maternal age, maternal education, risk status, clinic type and times that of adequate category and the odds for preterm birth in
utilization adequacy or content adequacy. the inadequate category was 2.34 times that of the adequate cat-
egory (Table 4). Although it did not reach statistical significance, the
finding suggested that women with adequate utilization were asso-
Results ciated with lower odds of preterm birth compared with those with
Respondents’ characteristics: users of ANC inadequate and intensive utilization. Having intensive utilization
A total of 522 records were analysed; characteristics of the women appeared to be associated with higher prevalence of preterm birth.
are presented in Table 1.
Preterm birth rate per 100 live births was ~7%, lower than offi- Adequacy of content and pregnancy outcomes
cial rate of 12% [11]. The lower rate was due to the role of first Based on the observed significance level, ANC content was asso-
level health facilities in which the extreme high-risk conditions ciated with preterm birth statistically. The odds of preterm birth in
including those susceptible to preterm birth were often referred and inadequate category were over three times that of adequate category
managed at hospital level. The mean gestational age of preterm (Table 4). ANC content did not appear to influence LBW.
births was 34 weeks (min 29, max 36). LBW was <13%, almost
similar to official rate of 11% [11]. Regarding timing of first visit,
50% of the preterm births had the first visit by 12 weeks, and 50% Discussion
after 12 weeks. Forty-five percent of the LBW had the first visit by We analysed adequacy of utilization and adequacy of content separ-
12 weeks and 55% after. ately, to evaluate ANC. This approach was more useful compared
with a composite measure that combined both utilization and con-
Table 1 Characteristics of the women (n = 522) tent as one variable. The composite measure cannot ascertain if the
adverse pregnancy outcomes were associated with utilization or con-
Characteristics n (%)
tent inadequacy, and thus, would be less specific in diagnosing gaps
Maternal age at first visit in care delivery. Instead, we used binary logistic regression which
≤19 11 (2.1%) included either utilization adequacy or content adequacy as predic-
20–34 439 (84.1%) tors (independent variables) of adequacy and pregnancy outcomes.
≥35 72 (13.8%)
Mean (SD) 28.7 (5.0)
Table 2 Distribution of women by adequacy of utilization and/or
Maternal education
content (n = 522)
Primary or no formal education 23 (4.4%)
Secondary 294 (56.3%) n (%)
Tertiary 193 (37.0%)
Unknown 12 (2.3%) Adequacy of utilization (APNCU-Malaysia index)
Parity Inadequate 107 (20.5)
Nulliparous 195 (37.4%) Adequate 85 (16.3)
Multiparous 327 (62.6%) Adequate-plus 330 (63.2)
Mean (SD) 1.2 (1.3) Total 522 (100.0)
Risk status Adequacy of content
Low-risk 375 (71.8%) Inadequate (<80%) 270 (51.7)
High-risk 147 (28.2%) Adequate (≥80%) 252 (48.3)
Clinic type by planned daily workload Total 522 (100.0)
<150 98 (18.8%) Adequacy of utilization * adequacy of content
150–300 247 (47.3%) Inadequate utilization * inadequate content 62 (11.9)
301–500 177 (33.9%) Inadequate utilization * adequate content 45 (8.6)
Pregnancy outcomes Adequate utilization * inadequate content 208 (39.8)
Preterm birth 36 (6.9%) Adequate utilization * adequate content 207 (39.7)
LBW 66 (12.7%) Total 522 (100.0)

Downloaded from https://academic.oup.com/intqhc/article-abstract/30/6/466/4953352

by guest
on 02 August 2018
 Evaluating antenatal care • Patient outcomes 469

Table 3 Difference of mean for antenatal care content score in % the case of Malaysia, the recommended initiation period is by 12
by antenatal care utilization level (n = 522) weeks of gestation [13]. It has been recognized that women who
had late initiation might not have the opportunity to benefit from
Mean antenatal P-value
screening tests, antenatal health advice or supported decision mak-
care content score
ing regarding choice of delivery [19, 20]. There is, however, limited
ANC utilization level <0.001 evidence to establish the right period for initiation of first visit [21,
Inadequate 76.3 22]. For example, the Confidential Enquiry into Maternal Deaths in
Adequate 74.6 the UK reveals that 43% of the deceased had their ANC appoint-
Adequate-plus 78.1 ment at or before 10 weeks of gestation, 31% at 11–12 weeks, and
Pairwise comparison
20% after 12 weeks (missing data 6%) [22]. In total, over 70% of
Inadequate and adequate-plus 0.048
women who died had their appointment by 12 weeks of gestation,
Inadequate and adequate 0.239
compared with 20% with booking after 12 weeks. Considering the
Adequate and adequate-plus <0.000
lack of evidence to determine the exact appropriate first antenatal
General linear model univariate full model analysis containing parity, risk visit, we decided to adhere to the concept of the long researched
level, clinic type, percentage of total visits attended by specific providers and APNCU index, which categorized initiation at or before 4 months
antenatal care utilization adequacy. gestation in the adequate category [6].
In Malaysia, indicators for ANC content measure single proced-
ure such as crude coverage of anti-tetanus toxoid (ATT) immuniza-
Table 4 Adequacy of antenatal care and pregnancy outcomes
tion, haemoglobin status at ±36 weeks gestation and number of
models
medical referrals (either to medical officer, family medicine specialist
ANC adequacy Adjusted OR* (95% CI) or hospital). Such indicators have limitation in reflecting quality of
Preterm birth (n = 36) LBW (n = 66)
ANC. For example, while coverage of ATT in this study has been
excellent (96%), this indicator is not able to reflect the overall qual-
ANC utilization ity of care as evidenced by the large proportion of women (52%)
Inadequate 2.34 (0.45–12.16) 0.73 (0.26–2.06) who received <80% of recommended routine care content.
Adequate 1.00 1.00 The slightly lower mean content score among women in the
Adequate-plus 3.27 (0.73–14.60) 1.36 (0.60–3.09) ‘adequate’ utilization category compared with the ‘adequate-plus’
ANC content
category was largely related to the scoring criteria of the ANC con-
Inadequate 3.69 (1.60–8.55) 1.17 (0.68–2.01)
tent (Supplementary Table S1). Majority of the items were adjusted
Adequate 1.00 1.00
for timing of initiation and birth, except for health education.
*
Full model binary logistic regression, odds ratios (ORs) for antenatal care Having higher utilization level (adequate-plus) implies higher num-
adequacy (utilization and content) were adjusted for maternal age, maternal ber of visits than routinely recommended; this presents more oppor-
education, risk status, clinic type and utilization or content. tunities for these women to receive additional health advice, than
women with adequate level of utilization. Nevertheless, utilization
level higher than recommended should not be encouraged especially
We also recognized the importance of including risk status of among low-risk pregnancies. Studies have concluded that ANC for
women as an independent variable, which facilitated analysis on women without risk could be provided with fewer visits [21, 23].
rational use of care according to need. Delivery of ANC should aim for completeness of care within the
In many developing settings, indicators related to ANC utiliza- recommended schedule.
tion often include timing of first visit and average number of total Pregnancy outcomes depend on a complex network of interac-
visits, which are analysed separately. The nature of these single indi- tions between care provided, the individuals concerned and the con-
cators, though commonly used, indicates only what it is defined; text in which they occur [24]. Therefore, the inclusion of only a few
they are not designed to provide comprehensive measurement on selected interventions may lead to a false linear relationship between
quality of care. The indicator, timing of first visit, measures only the interventions and outcomes. To overcome these limitations, our
what is intended and does not provide information after the first vis- study included the entire recommended routine care of the national
it. Likewise, ‘average number of total ANC visits’ provides little guidelines.
benefit in assessing adequacy of utilization when the initiation peri- An advantage of the content adequacy assessment tool is that it
od is not known. For example, a multiparous could be seen seven enables modification, to include or omit other interventions accord-
times throughout her 40-week pregnancy, which is considered ing to the study context. It accounts for under-provision of care con-
adequate according to the recommended number of ANC visits for tent but not over-provision. We, therefore, proposed to analyse
uncomplicated multipara [13]. However, if it was not known that selected interventions separately, in particular, those with high-cost
the woman had her first ANC visit at 20 weeks gestation, it would implication, for example, the number of ultrasounds a woman
be inappropriate to classify the woman as having adequate utiliza- received.
tion because she fulfilled the recommended number of ANC visits. The proportion of ‘adequate-plus’ category in our study was
The scenario highlights the limitation in using these two indicators higher than that of a study in the USA, which recorded 30% classi-
separately and reiterates the importance of using indicators like the fied as ‘adequate-plus’ and 45% adequate utilization [25]. The
APNCU index to assess ANC utilization. recommended ANC schedule of the USA is higher than that of the
The finding demonstrated the absence of significant differences Malaysian schedule; therefore applying the US standard may result
in terms of pregnancy outcomes between the women who had the in a lower proportion of ‘adequate-plus’. However, national stan-
first ANC visit by 12 weeks and after 12 weeks. Monitoring gesta- dards are country-specific, according to availability of technical,
tional age of first visit aimed to promote early initiation of care. In human and financial resources, and affordability. Therefore, adequacy

Downloaded from https://academic.oup.com/intqhc/article-abstract/30/6/466/4953352

by guest
on 02 August 2018
 470
of care is specific to national standards, of which each operates within
a substantially different resource envelope.
Our study found that over half of the women had <80% of the
recommended routine ANC content documented in their antenatal
records (52% vs 48%). A previous study agrees with this finding
(70% vs 30%) [26]. The results from this present study appear to
indicate better adherence to recommended routine content com-
pared with the other study, in terms of proportion of women with
≥80% of ANC content (48% vs 30%). However, there are differ-
ences in assessment criteria. For example, the assessment criteria for
maternal weight in the American study required that maternal
weight be measured at every visit, whereas the assessment criteria
for maternal weight in this present study considered the recom-
mended number of visits. If the total visits are less than the recom-
mended schedule, evidence of weight taken at each visit is accepted
as having met the scoring criteria (Supplementary Table S1).
Another study that used the APNCU index and its variants,
found that the odds for preterm births were higher among those
who had intensive (adequate-plus) and inadequate utilization than
adequate utilization [25]. Although it did not reach statistical signifi-
cance, women who had adequate-plus utilization appeared to be
associated with higher prevalence of preterm births in our study. A
significantly larger proportion of high-risk women were shown to
have intensive utilization and poorer pregnancy outcomes [18]. This
explained the association between intensive utilization and preterm
birth since high-risk women, in general, were scheduled for frequent
monitoring visits.
ANC content adequacy was significantly associated with preterm
birth but did not appear to influence LBW. This is consistent with a
study which reported that low adherence to routine ANC content
(<80% of recommended routine practice) was associated with
increased odds of preterm birth (adjusted odds ratios (aOR) = 1.8,
95% confidence interval (CI) = 1.0–3.4), but had no effect on LBW
(aOR = 1.0, 95% CI = 0.5–1.9) [26]. As discussed elsewhere,
women who deliver preterm would have lesser opportunities to
receive routine care which might have contributed to improved out-
comes [17]. However, in this study, the scoring criteria for the ANC
content considered the gestational age of birth where relevant,
except for health education. Furthermore, the mean gestational age
of preterm births was 34 weeks, wherein health education topics
assessed in this study could have already been provided prior.
The limitations associated with the study design have been dis-
cussed elsewhere [17, 18]. Though facility-based routine data have
high relevance to improving quality and outcomes of care, we recog-
nize the challenges associated with data extraction in a paper-based
medical record system. Nevertheless, in the face of stagnating pro-
gress in pregnancy outcomes, we believe initiation of such assess-
ment is crucial. The effect between ANC adequacy and pregnancy
outcomes would have reached statistical significance if a larger sam-
ple size was enroled. The effects of independent variables on preg-
nancy outcomes were not explored since this paper specifically deals
with quality of ANC and pregnancy outcomes.
Our study shows inadequate ANC content is associated with
higher prevalence of preterm birth and provides suggestive evidence
that inadequate utilization of ANC increases the risk of preterm
birth. The study demonstrates relevance of using both utilization
and content assessment in evaluating quality of ANC. The findings
demonstrated logical consistency as compared with other studies. In
summary, the methods could be reviewed and may allow for com-
prehensive evaluation of technical performance, thereby, facilitating
improving quality and informing policy formulation. The initiation
Downloaded from https://academic.oup.com/intqhc/article-abstract/30/6/466/4953352
by guest
on 02 August 2018
Yeoh et al.
and continuous refinement of the evaluation methods will bring
about an impetus to address gaps in ANC.
Supplementary material
Supplementary material is available at International Journal for Quality in
Health Care online.
Acknowledgements
The authors wish to acknowledge the contribution of Dr Ophelia Mendoza,
Adjunct Prof of the Department of Epidemiology and Biostatistics, College of
Public Health, University of the Philippines, for the advice on sampling and
sample size estimate; Prof Dr Karuthan Chinna and Prof Dr Sanjay Rampal
of Department of Social and Preventive Medicine, University of Malaya, for
the advice on data analysis; and the Department of Health of Selangor, for
allowing the study to be conducted.
Authors’ contribution
Conceived or designed the work: all. Acquired, analysed or interpreted data
for the work: all. Drafted the work or revised it critically for important intel-
lectual content: all. Approved the final version to be published: all. Agreed to
be accountable for the work: all.
Funding
This work was supported by Mediconsult Sdn. Bhd. to PLY for a PhD study.
The funder had no role in study design, data collection and analysis, decision
to publish or preparation of the manuscript.
References
1. Alexander GR, Kotelchuck M. Quantifying the adequacy of prenatal
care: a comparison of indices. Public Health Rep 1996;111:408–18. dis-
cussion 419.
2. Alexander GR, Kotelchuck M. Assessing the role and effectiveness of pre-
natal care: history, challenges, and directions for future research. Public
Health Rep 2001;116:306–16.
3. Bloch JR, Dawley K, Suplee PD. Application of the Kessner and
Kotelchuck prenatal care adequacy indices in a preterm birth population.
Public Health Nurs 2009;26:449–59.
4. Beeckman K, Fred Louckx F, Masuy-Stroobant G et al. The development
and application of a new tool to assess the adequacy of the content and
timing of antenatal care. BMC Health Serv Res 2011;11:213.
5. Larson E, Vail D, Mbaruku GM et al. Beyond utilization: measuring
effective coverage of obstetric care along the quality cascade. Int J Qual
Health Care 2017;29:104–10.
6. Kotelchuck M. An evaluation of the Kessner Adequacy of Prenatal Care
Index and a proposed Adequacy of Prenatal Care Utilization Index. Am J
Public Health 1994;84:1414–20.
7. WHO. WHO Recommended Interventions for Improving Maternal and
Newborn Health, 2nd edn. Geneva: WHO Department of Making
Pregnancy Safer, 2009.
8. Pittrof R, Campbell OMR, Filippi VGA. What is quality in maternity
care? An international perspective. Acta Obstet Gynecol Scand 2002;81:
277–83.
9. Bloom SS, Lippeveld T, Wypij D. Does antenatal care make a difference
to safe delivery? A study in urban Uttar Pradesh, India. Health Policy
Plan 1999;14:38–48.
10. Ministry of Health Malaysia (2009) Annual Report 2008 MOH
Malaysia.
11. WHO. World Health Statistics 2014. Geneva: WHO, 2014.
 Evaluating antenatal care • Patient outcomes
12. Ministry of Health Malaysia (2012) Annual Report Family Health 2010
—Health Information and Management System. In: Health Informatics
Centre PaDD, editor.
13. Ministry of Health. Perinatal care manual: Section 2 Antenatal care, 2nd
edn. Kuala Lumpur: Division of Family Health Development, Ministry of
Health Malaysia, 2010.
14. Ministry of Health Malaysia (2010) Annual Report 2009 MOH
Malaysia.
15. WHO. World Health Statistics 2012. Geneva: WHO, 2012.
16. Ministry of Health Malaysia (2014) Health indicators 2014—Indicators
for monitoring and evaluation of strategy Health For All. In: Health
Informatics Centre PDDM, editor.
17. Yeoh PL, Hornetz K, Ahmad Shauki NI et al. Assessing the extent of
adherence to the recommended antenatal care content in Malaysia: room
for improvement. PLoS One 2015;10:e0135301.
18. Yeoh PL, Hornetz K, Dahlui M. Antenatal care utilisation and content
between low-risk and high-risk pregnant women. PLoS One 2016;11:
e0152167.
19. Baker EC, Rajasingam D. Using Trust databases to identify predictors of late
booking for antenatal care within the UK. Public Health 2012;126:112–6.
20. Agha S, Tappis H. The timing of antenatal care initiation and the content
of care in Sindh, Pakistan. BMC Pregnancy Childbirth 2016;16:190.
Downloaded from https://academic.oup.com/intqhc/article-abstract/30/6/466/4953352
by guest
on 02 August 2018
471
21. Villar J, Ba’aqeel H, Piaggio G et al. WHO antenatal care randomised
trial for the evaluation of a new model of routine antenatal care. Lancet
2001;357:1551–64.
22. Knight M, Tuffnell D, Kenyon S et al (eds). on behalf of MBRRACE-UK.
Saving Lives, Improving Mothers’ Care—Surveillance of Maternal Deaths
in the UK 2011–13 and Lessons Learned to Inform Maternity Care from
the UK and Ireland Confidential Enquiries into Maternal Deaths and
Morbidity 2009–13. Oxford: National Perinatal Epidemiology Unit,
University of Oxford, 2015.
23. Carroli G, Villar J, Piaggio G et al. WHO systematic review of rando-
mised controlled trials of routine antenatal care. Lancet 2001;357:
1565–70.
24. Enkin MW, Glouberman S, Groff P et al. Beyond evidence: the complex-
ity of maternity care. Birth 2006;33:265–9.
25. VanderWeele T, Lantos J, Siddique J et al. A comparison of four prenatal
care indices in birth outcome models: comparable results for predicting
small-for-gestational-age outcome but different results for preterm birth
or infant mortality. J Clin Epidemiol 2009;62:438–45.
26. Handler A, Rankin K, Rosenberg D et al. Extent of documented adher-
ence to recommended prenatal care content: provider site differences and
effect on outcomes among low-income women. Matern Child Health J
2012;16:393–405.