21/10/2016
1
Antenatal care use in urban areas in two European
countries: Predisposing, enabling and pregnancy-
related determinants in Belgium and the
Netherlands
Jana Vanden Broeck, dr Esther Feijen-de Jong, dr Trudy Klomp, prof dr Koen Putman,
prof dr Katrien Beeckman
Antenatal care is important
∎ Role of antenatal care
∎ Physical importance
∎ Prevention – treatment
∎ Psychological importance
∎ Information – preparation
∎ Fetal origins hypothesis (Barker, 2002-2007-2009)
2 Prenatale zorg in Brussel
21/10/2016
2
Disparities in antenatal care use
titel 3 Andersen and Newman’s Health Behaviour model, 1973
Societal Determinants Health Services System
Technology
Norms
Resources
Organisation
Individual determinants
Health Care Utilisation
Predisposing
Enabling
Need
Type
Purpose
Unit of Analysis
Disparities in antenatal care use
Individual determinants related to inadequate Antenatal care use:
Predisposing: Maternal Age, being single, low educational
level, lack of paid job, foreign ethnic background or origin,
poor language proficiency, lack of support social network, lack
of knowledge HCS, etc
Enabling: absence of health insurance, hospital type, care
provider’s characteristics, etc.
Need component: parity, unplanned pregnancy, previous
preterm birth, discontinuity of care, smoking, etc.
Baker et al, 2012; Feijen de Jong, 2012; Heaman, 2013; Beeckman et al, 2013; Choté et al, 2011 4
Individual determinants
Predisposing
Enabling
Need
21/10/2016
3
Disparities in antenatal care use
Societal Determinants:
Country specific organisation, eg referral system between health care
providers
Health Services System:
Resources: eg Number of care providers available
Need for comparative research in several countries with varying
antenatal health care arrangements as these might explain
differences in the effects of individual determinants on ANC use
titel 5 21-10-2016
Societal Determinants Health Services System
Technology
Norms
Resources
Organisation
Measuring antenatal care use
Number of visits
Number of visits in relation to pregnancy
duration: eg APNCU index (Alexander and Kotelchuck, 2001)
Measures including content of care
Content and Timing of Care in Pregnancy tool
(CTP) (Beeckman et al, 2011)
titel 6 21-10-2016
Health Care Utilisation
Type
Purpose
Unit of Analysis
21/10/2016
4
Measuring antenatal care use
Content and Timing of Care in Pregnancy tool
Literature relevance interventions in pregnancy
Congruences in (inter)national guidelines for routine care
in pregnancy
CTP reflects a minimal care package
recommended in every pregnancy regardless risk
status or parity
Timing of initiation of care
Number and appropriate timing of three interventions in
pregnancy
(Beeckman et al, 2011)
titel 7 21-10-2016
U: Ultrasounds BP: Blood Pressure BS: Blood Sample T: Trimester
YES
Prenatale zorg in Brussel 8 21-10-2016
1. Initiation of care before 14 weeks of gestation
NO = Inadequate
2. Number of interventions:
3. Timing of interventions:
Not all interventions occurred according the
time table = Sufficient
All interventions occurred according the time
table = Appropriate
Minimum number per trimester
U BP BS
T1 1 1 1
T2 1 2 0
T3 0 3 1
U BP BS
> 2 en > 6 en > 2 )
U BP BS
(< 4 en < 11 en < 4 )
YES= Inadequate
NO= Intermediate
NO
YES
21/10/2016
5
Aim of the study
Comparison of Antenatal care utilisation
in Belgium and the Netherlands
Two countries with different health care
system
Netherlands: primary care midwives, women
receive fixed remunerations
Belgium: most women access an obstetrician
directly, payment on a fee-for-service basis
Gain insight in predisposing, enabling
and pregnancy related determinants
titel 9 21-10-2016
Methods
Secondary data analysis
Belgium: prospective observational study in the
Brussels Metropolitan Region
- Recruitment:
Consecutive (April - July 2008);
9 Brussels hospitals;
Inclusion criteria: Women aged >18; speaking (Dutch, French,
English, Turkish or Arabic); reachable by phone; living in the
Brussels Metropolitan Region, no major deseases;
maximum3th visit
- Data Collection:
Structured questionaires: beginning and end; diary
Selfregistration of each visit, structured
Follow-up: bimonthly telephone call
(Beeckman et al, 2011)
titel 10 21-10-2016
21/10/2016
6
Methods
Secondary data analysis
Netherlands: Data Primary care Delivery
- Recruitment:
purposive sampling of 20 midwifery practices spread over the
country,
consecutive inclusion of all women attending the practice
between 2009-2010, eligible when understanding Dutch,
English, Turkish or Arabic
- Data collection:
two questionnaires on demographics and ultrasound scans
(begin and end),
data on ANC was extracted from the electronic records
(Mannien et al, 2012)
titel 11 21-10-2016
Methods: Pooled dataset
- Adult women
- Residing an urban
region
- Low-risk onset of
pregnancy
N= 632 out of
7907
Whole sample
N=333
titel 12 * Exact matching on household income, occupation and educational level, Hot Deck method was used
Random matched *
sample n= 321 from
each study
21/10/2016
7
Methods
Individual determinants available in both
studies
Predisposing: age, marital status, educational
level, occupational status and current
nationality, education of partner, current
nationality of partner, country of residence
Enabling: equivalent income, health insurance
cover, additional health insurance cover
Need/pregnancy related: parity, wish for
pregnancy planned pregnacy, continuity of
care, attendance of antenatal informtion
classes
titel 13 21-10-2016
Methods
Statistical analysis
Descriptive statistics
Comparison of study population in both
datasets: Chi² - analysis
Logisitic ordinal regression analyses,
likelihood of having a better care trajectory
(higher CTP classification) while controlling of
the remaining significant characteristics.
Proportional odds assumption
titel 14 21-10-2016
21/10/2016
8
Results, predisposing characteristics
Brussels Metropolitan Region The Netherlands
(N = 321) (N = 321) Chi2
N (column %) N (column %) (p-value)
Age (years) < =20 4 (1.2) 4 (1.2) 1.000
21-35 264 (82.2) 264 (82.2)
> 35 53 (16.5) 53 (16.5)
Marital status Co-habiting or married 291 (90.7) 313 (97.5) .000
Single 30 (9.3) 8 (2.5)
Educational level Up to secondary 188 (58.6) 188 (58.6) 1.000
Tertiary 133 (41.4) 133 (41.4)
Occupational status Employed 149 (46.4) 270 (84.1) .000
Unemployed 172 (53.6) 51 (15.9)
Nationality Belgian/Dutch 184 (57.3) 291 (90.7) .000
All other nationalities 137 (42.7) 30 (9.3)
titel 15 21-10-2016
Results, predisposing characteristics
Brussels Metropolitan Region The Netherlands
(N = 321) (N = 321) Chi2
N (column %) N (column %) (p-value)
Educational level partner No partner 30 (9.3) 8 (2.5) .001
Up to secondary 163 (50.8) 171 (53.3)
Tertiary 128 (39.9) 142 (44.2)
Nationality of the partner
No partner 30 (9.3) 8 (2.5) .000
Belgian/Dutch 170 (53.0) 271 (84.4)
All other nationalities 121 (37.7) 42 (13.1)
titel 16 21-10-2016
21/10/2016
9
Results, enabling characteristics
Brussels Metropolitan Region The Netherlands
(N = 321) (N = 321) Chi2
N (column %) N (column %) (p-value)
Equivalent incomea Low 92 (28.7) 20 (6.2) .000
Moderate 151 (47.0) 300 (93.5)
High 78 (24.3) 1 (0.3)
Health insurance coverage
Yes 302 (94.1) 321 (100.0) .000
No 19 (5.9) 0 (0.0)
Additional health insurance coverage
Yes 151 (47.0) 280 (87.2) .000
No 170 (53.0) 41 (12.8)
a∑ incomes in the household/(1 + (x*0.5) + (y*0.3)) (x: number of adults living in the same household, y: number of children under the age
of 18 years living in the same household [modified OECD scale] 17
Results, pregnancy related
characteristics
Brussels Metropolitan Region The Netherlands
(N = 321) (N = 321) Chi2
N (column %) N (column %) (p-value)
Parity Primiparae 121 (37.7) 163 (50.8) .001
Multiparae 200 (62.3) 158 (49.2)
Wish for pregnancyb Wanted pregnancy 308 (96.3) 320 (99.7) .002
Unwanted pregnancy 12 (3.8) 1 (0.3)
Planned pregnancy Yes 250 (77.9) 262 (81.6) .239
No 71 (22.1) 59 (18.4)
COCc < 50 % 238 (74.1) 225 (70.1) .253
> =50 % 83 (25.9) 96 (29.9)
Attending antenatal information courses
Yes 71 (22.1) 167 (52.0) .000
No 250 (77.9) 154 (48.0)
Continuity of Care index: COC=∑n 2 j −n/n(n−1) 18 21-10-2016
21/10/2016
10
Results
titel 19 21-10-2016
9,7 5,6
8,1
6,2
36,8
29,9
45,5
58,3
B RUS S ELS METRO PO LITA N REG IO N URB A N D UTCH REG IO NS
ANTENATAL CARE USE (CTP)
Inadequate Intermediate Sufficient Appropriate
p = 0,009, Chi² test
Results
Relation between study sample
characteristics and antenatal care
utilisation
Chi² test
titel 20 21-10-2016
21/10/2016
11
Results, predisposing characteristics
Antenatal care utilisation classified by the CTP tool P value
Inadequate Intermediate Sufficient Appropriate χ2 test
(N = 49) (N = 46) (N = 214) (N = 333)
N (row %) N (row %) N (row %) N (row %)
Age (years) ≤20 0 0 6 (75.0) 2 (25.0) 0.32(a)
21–35 41 (7.8) 40 (7.6) 172 (32.6) 275 (52.1)
>35 8 (7.5) 6 (5.7) 36 (34.0) 56 (52.8)
Marital status Co-habiting or married
44 (7.3) 45 (7.5) 197 (32.6) 318 (52.6) 0.14(a)
Single 5 (13.2) 1 (2.6) 17 (44.7) 15 (39.5)
Educational level Employed 20 (4.8) 26 (6.2) 120 (28.6) 253 (60.4) <0.001
Unemployed 29 (13.0) 20 (9.0) 94 (42.2) 80 (35.9)
Occupational status
Up to secondary 35 (9.3) 33 (8.8) 139 (37.0) 169 (44.9) <0.001
Tertiary 14 (5.3) 13 (4.9) 75 (28.2) 164 (61.7)
Continuity of Care index: COC=∑n 2 j −n/n(n−1) 21 21-10-2016
Results, predisposing characteristics
Antenatal care utilisation classified by the CTP tool P value
Inadequate Intermediate Sufficient Appropriate χ2 test
(N = 49) (N = 46) (N = 214) (N = 333)
N (row %) N (row %) N (row %) N (row %)
Nationality Belgian/Dutch 29 (6.1) 36 (7.6) 149 (31.4) 261 (54.9) 0.009
All other nationalities
20 (12.0) 10 (6.0) 65 (38.9) 72 (43.1)
Educational level partner
No partner 5 (13.2) 1 (2.6) 17 (44.7) 15 (39.5) <0.001
Up to secondary 33 (9.9) 30 (9.0) 120 (35.9) 151 (45.2)
Tertiary 11 (4.1) 15 (5.6) 77 (28.5) 167 (61.9)
Nationality of the partner
No partner 5 (13.2) 1 (2.6) 17 (44.7) 15 (39.5) 0.003
Belgian/Dutch 27 (6.1) 29 (6.6) 133 (30.2) 252 (57.1)
All other nationalities
17 (10.4) 16 (9.8) 64 (39.3) 66 (40.5)
Region Brussels Metropolitan
31 (9.7) 26 (8.1) 118 (36.8) 146 (45.5) 0.009
Urban Dutch regions
18 (5.6) 20 (6.2) 96 (29.9) 187 (58.3) Continuity of Care index: COC=∑n 2 j −n/n(n−1) 22 21-10-2016
21/10/2016
12
Results, enabling characteristics
Antenatal care utilisation classified by the CTP tool P value
Inadequate Intermediate Sufficient Appropriate χ2 test
(N = 49) (N = 46) (N = 214) (N = 333)
N (row %) N (row %) N (row %) N (row %)
Equivalent incomea
Low 17 (15.2) 9 (8.0) 51 (45.5) 35 (31.3) <0.001
Moderate 29 (6.4) 33 (7.3) 141 (31.3) 248 (55.0)
High 3 (3.8) 4 (5.1) 22 (27.8) 50 (63.3)
Health insurance coverage
Yes 46 (7.4) 46 (7.4) 202 (32.4) 329 (52.8) 0.008
No 3 (15.8) 0 (0.0) 12 (63.2) 4 (21.1)
Additional health insurance coverage
Yes 24 (5.6) 29 (6.7) 130 (30.2) 248 (57.5) <0.001
No 25 (11.8) 17 (8.1) 84 (39.8) 85 (40.3)
Continuity of Care index: COC=∑n 2 j −n/n(n−1) 23 21-10-2016
Results, pregnancy related
characteristics
Antenatal care utilisation classified by the CTP tool P value
Inadequate Intermediate Sufficient Appropriate χ2 test
(N = 49) (N = 46) (N = 214) (N = 333)
N (row %) N (row %) N (row %) N (row %)
Parity Primiparae 19 (6.7) 16 (5.6) 84 (29.6) 165 (58.1) 0.042
Multiparae 30 (8.4) 30 (8.4) 130 (36.3) 168 (46.9)
Wish for pregnancyb
Wanted pregnancy 49 (7.8) 44 (7.0) 210 (33.4) 325 (51.8) 0.51(a)
Unwanted pregnancy
0 (0.0) 2 (15.4) 4 (30.8) 7 (53.8)
Planned pregnancy
Yes 35 (6.8) 34 (6.6) 161 (31.4) 282 (55.1) 0.013
No 14 (10.8) 12 (9.2) 53 (40.8) 51 (39.2)
COCc <50 % 42 (9.1) 39 (7.8) 158 (34.1) 227 (49.0) 0.041
≥50 % 7 (3.9) 10 (5.6) 56 (31.3) 106 (59.2)
Attending antenatal information courses
Yes 11 (4.6) 7 (2.9) 72 (30.3) 148 (62.2) <0.001
No 38 (9.4) 39 (9.7) 142 (35.1) 185 (45.8)
Continuity of Care index: COC=∑n 2 j −n/n(n−1) 24 21-10-2016
21/10/2016
13
Results
Relation between study sample
characteristics and antenatal care
utilisation
Ordinal regression analyses, adjusted OR for
being assigned into a higher CTP category
titel 25 21-10-2016
62,2
45,8
59,2
49
58,3
45,5
61,7
44,9
52,6
35,9
A N T E N A T A L C L A S S E S
N O A N T E N A T A L C L A S S E S
C O C > = 5 0 %
C O C < 5 0 %
U R B A N D U T C H R E G I O N S
B R U S S E L S M E T R O P O L I T A N
T E R T I A R Y E D U C A T I O N
U P T O S E C O N D A R Y E D U C A T I O N
E M P L O Y E D
U N E M P L O Y E D
PERCENTAGE OF WOMEN IN THE CTP APPROPRIATE GROUP
AdjOR 0,60 (0,43-0,82)
AdjOR 0,90 (0,64-1,26)
AdjOR 0,67 (0,47-0,94)
Results
titel 26 21-10-2016
AdjOR 0,45 (0,34-0,70)
AdjOR 0,60 (0,42-0,84)
21/10/2016
14
Discussion
Urban Dutch women more appropriate ANC ~
women from Brussels Metropolitan Region
Multivariate analysis do not indicate region itself
as determinant of ANC use
Adequate content and timing of ANC is associated
with higher educational level, employed status,
higher continuity of care and attendance of
antenatal classes
titel 27 21-10-2016
Most important determinants of ANC
Educational level
General health knowledge (Chote et al, 2012)
Health literacy (Chote et al, 2012)
Kwoledge and skills aquired through
education (Solar et al, 2005)
better access to information
receptiveness to health education messages
Enables access and communication with care
providers
titel 28 21-10-2016
21/10/2016
15
Most important determinants of ANC
Employment
Social network more extented? (Deri, 2005; Lin, 1999)
Information and encouragement might
stimulate ANC use (Deri, 2005; Lin, 1999)
Continuity of care provider
Positive effect on ANC use, irrespective of type
of caregiver
Attenting antenatal information classes
Non-attenders, less convinced of imporatance
and need for ANC (Murphy, 2010)
Non-attenders, often from vulnerable groups
(Fabian et al, 2004)
titel 29 21-10-2016
No enabling determinant
None of the enabling determinants were
retained in final model
Income does not play a role
Universal coverage, including ANC in BE and
NL, might be explaination
titel 30 21-10-2016
21/10/2016
16
Study limitations
Variables needed to be equally
operationalised
Missing of Origin or Ethnicity !
Important in original studies as well as international
literature
More variables on health care system =
valuable
Main care provider, reimbursement system
Women that do not seek care are not
included
titel 31 21-10-2016
Tackling inequalities in ANC use
Universal covarage seems insufficient to
offset disparities in ANC use in both
countries (Simoes et al, 2006)
The focus on educational level and
employement status as structural aspects of
society have a low mutability in the
promotion of equal access (cfr. Andersen, 1994)
titel 32 21-10-2016
21/10/2016
17
Tackling inequalities in ANC use
Better investments are
- Promotion of health literacy (Parker et al, 2003)
- Promotion of knowledge on health care system (Parker et al, 2003)
through the educational sytem:
primary - secundary school, integration courses
- Training of health professionals: communication skills to
adapt to health literacy level of the patient (Parker et al, 2003)
- Stimulation of continuity in maternal health care models
- Promotion of antenatal classes, eg free to vulnerable women
titel 33 21-10-2016
Midwifery in international perspective
Migration and Poverty
Migration, important determinant of ANC use in the original
studies
Non-western women have double the risk to receive inadequate
ANC compared to Western Non-Dutch or Native Dutch women
(adjOR 2,05 (95%CI 1,48-2,85)) (Feijen-de Jong et al, 2015)
Maghreb Women have less than half the chance of having an
appropriate care trajectory compared to Belgian women or
women with another origin (adjOR 0,38 (95%CI 0,22-0,66))
(Beeckman et al, 2013)
Employment status and educational level are determinanst of
ANC use
More important than the country a women lives in
Unemployment and low educational level contribute to poverty
titel 34 21-10-2016
21/10/2016
18
Thank you for your attention
Published in BMC Health Services Research, (2016) 16:337
21-10-2016 35