The authors have declared that no competing interests exist.
Conceived and designed the experiments: J. Stephenson DP GB BH AC OO PP J. Shawe. Analyzed the data: DP AC BH. Contributed reagents/materials/analysis tools: J. Stephenson DP GB BH AC OO PP J. Shawe. Wrote the paper: J. Stephenson DP GB AC OO PP J. Shawe.
To determine the extent to which women plan and prepare for pregnancy.
Cross-sectional questionnaire survey of pregnant women attending three maternity services in London about knowledge and uptake of preconception care; including a robust measure of pregnancy planning, and phone interviews with a range of health care professionals.
We recruited 1173/1288 (90%) women, median age of 32 years. 73% had clearly planned their pregnancy, 24% were ambivalent and only 3% of pregnancies were unplanned. 51% of all women and 63% of those with a planned pregnancy took folic acid before pregnancy. 21% of all women reported smoking and 61% reported drinking alcohol in the 3 months before pregnancy; 48% of smokers and 41% of drinkers reduced or stopped before pregnancy. The 51% of all women who reported advice from a health professional before becoming pregnant were more likely to adopt healthier behaviours before pregnancy [adjusted odds ratios for greatest health professional input compared with none were 2.34 (95% confidence interval 1.54–3.54) for taking folic acid and 2.18 (95% CI 1.42–3.36) for adopting a healthier diet before pregnancy]. Interviews with 20 health professionals indicated low awareness of preconception health issues, missed opportunities and confusion about responsibility for delivery of preconception care.
Despite a high level of pregnancy planning, awareness of preconception health among women and health professionals is low, and responsibility for providing preconception care is unclear. However, many women are motivated to adopt healthier behaviours in the preconception period, as indicated by halving of reported smoking rates in this study. The link between health professional input and healthy behaviour change before pregnancy is a new finding that should invigorate strategies to improve awareness and uptake of pre-pregnancy health care, and bring wider benefits for public health.
The period before conception is increasingly regarded as important for the health of pregnant women and future generations. Successive reports from the Centre for Maternal and Child Enquiries
There is a fair degree of consensus among expert bodies and in professional guidelines about what preconception care should entail, particularly in the USA. It includes folic acid supplementation for all women to prevent neural tube defects
However, despite the importance attributed to good pre-pregnancy care, there is little understanding of women's behaviour or the information they acquire in preparation for pregnancy, or how this relates to uptake of care or interaction with health care professionals. Government policy in UK and USA aims to reduce perinatal morbidity and mortality by promotion of preconception care
The survey was approved by the National Research Ethics Service, NRES Committee London- Bromley (REC reference 11/LO/0881). The approval was given as part of a larger study of preconception health and care in England. In accordance with standard practice approved by the research ethics committee, women gave consent by opting to complete the questionnaire. Women who agreed to follow up gave explicit signed consent.
The antenatal survey was conducted between November 2011 and May 2012 in the maternity services of three North London Hospitals, which were selected to enable women from diverse ethnic and socioeconomic backgrounds to participate. Women attending these maternity services represent a mix of both low and high risk pregnancies.
Women were approached by trained researchers. They were given an information leaflet about the project and the consent process, and invited to consent to completing the antenatal questionnaire, and being contacted for follow-up questionnaire or interview. Women who did not wish to be followed-up were invited to complete the baseline questionnaire. For practical reasons, the recruitment process varied by site, but the aim in all three hospitals was to recruit women early in pregnancy to reduce recall bias and to recruit from both low and high risk clinics. Women filled in a self-completion pen-and-paper questionnaire while they were waiting for their appointment. The data was entered onto computer by a commercial data entry company (Abacus).
We carried out a literature review to explore themes and topics that should be covered in the questionnaire and examined preconception care questionnaires used in the Southampton Women's Survey
The questionnaire asked whether respondents had visited a health professional (GP, nurse, midwife, consultant, pharmacist, family planning or antenatal professional) to obtain advice on getting pregnant; whether they had accessed information by any means about folic acid and vitamin supplements and a list of eleven other preconception health behaviours (healthy diet, healthy weight, alcohol, smoking, immunisations, recreational drugs, STIs, dental checks, caffeine, stopping contraception/fertility advice) from a health professional, from family or friends, or other sources including finding out for themselves. Health professional input was categorised on three levels: none; visiting a health professional and receiving advice on any of the eleven preconception health behaviours listed above
We determined the sample size by selecting at least 80% power to detect differences that we consider of practical importance at the 5% significance level in the prevalence of a range of key outcomes (e.g. folic acid consumption) where the prevalence may vary from low (e.g. 5–10%) to more common (e.g. 50%). Specifically we noted that a sample size of 948 participants provides 80% power to detect as significant a difference in the prevalence of an outcome between two groups of equal size (e.g. grouped by ethnicity or other factors) of 5% versus 10%, and 824 participants provides 80% power for a difference of 45% versus 55%. We therefore set a minimum sample size of 1000 participants so as to have more than 80% power for important differences across a range of outcomes of interest.
The chi-squared test was used to test associations between participant characteristics and measures of information acquired, professional input, behaviour change and folic acid consumption. All factors are treated as categorical and the categorisations used for the tests are those presented in the tables. Logistic regression was used to calculate unadjusted and adjusted odds ratios for a range of behaviour change outcomes and folic acid consumption, which are presented with 95% confidence intervals. The key explanatory factor investigated in the regressions is level of health professional input considered in 3 categories, whilst other participant characteristics are viewed as potential confounders. The set of potential confounders adjusted for to generate adjusted odds ratios is common to all outcomes and is all those factors found significantly associated with any of the behaviour change outcomes or with folic acid consumption in univariate analysis. Adjustment for age and LMUP score is made treating these as continuous factors, whilst other factors are treated as categorical with categories as presented in the tables.
We aimed to recruit a purposive sample of 20 health professionals from all nine governmental regions of England working in general practice, obstetrics & gynaecology, midwifery and sexual & reproductive health. All participants were contacted by email and invited to either face-to-face or telephone interview. The interviews explored how policy, guidelines and recommendations are implemented in day-to-day practice and identified perceived barriers. We used a topic guide consisting of a series of questions about what constitutes preconception care, the nature of current provision in their services, who should have responsibility for providing preconception care, to whom, the perceived barriers to service delivery and how the current situation could be improved. The interviews were carried out by four researchers between August 2011 and July 2012. They were tape-recorded and transcribed
We recruited 1173 of 1288 women invited to take part at the three study sites, giving an overall response rate of 90% (86%, 91% and 94% at each of the three hospitals).
Characteristic | % (n) | Extent of information acquired (any source) about 11 topics on preconception health | Level of health professional input before pregnancy | ||||
All | 34 (404) | 42 (487) | 24 (282) | 49(573) | 29(344) | 21 (250) | |
Age | |||||||
28 (288) | 41 (118) | 40 (116) | 19 (54) | 56(161) | 26 (74) | 18 (52) | |
41 (419) | 32 (132) | 42 (178) | 26 (109) | 48(202) | 29(122) | 22 (94) | |
31 (324) | 31 (102) | 44 (142) | 25 (80) | 48(153) | 30 (98) | 22 (71) | |
Ethnic group | |||||||
68 (705) | 32 (223) | 44 (307) | 25 (175) | 49 (346) | 29(203) | 22 (154) | |
6 (58) | 40 (23) | 40 (23) | 21 (12) | 53 (31) | 29 (17) | 17 (10) | |
12 (120) | 33 (39) | 43 (51) | 25 (30) | 45 (54) | 32 (38) | 23 (28) | |
10 (101) | 51 (52) | 31 (31) | 18 (18) | 55 (54) | 24 (24) | 20 (20) | |
6 (59) | 25 (15) | 51 (30) | 24 (14) | 54 (32) | 25 (15) | 20 (12) | |
Place of birth | |||||||
52 (539) | 33 (177) | 44 (238) | 23 (124) | 46 (245) | 32(171) | 22 (120) | |
48 (493) | 34 (167) | 41 (202) | 25 (124) | 54 (266) | 26(127) | 20 (99) | |
Employment status | |||||||
68 (714) | 30 (214) | 44 (313) | 26 (187) | 51 (365) | 28(196) | 21 (151) | |
10 (101) | 42 (42) | 36 (36) | 23 (23) | 40 (40) | 33 (33) | 27 (27) | |
19 (194) | 40 (77) | 41 (80) | 19 (37) | 46 (88) | 31 (60) | 23 (44) | |
4 (39) | 41 (16) | 49 (19) | 10 (4) | 54 (21) | 33 (13) | 13 (5) | |
Highest educational achievement | |||||||
66 (682) | 30 (205) | 45 (306) | 25 (171) | 49 (336) | 29(199) | 21 (144) | |
18 (183) | 37 (68) | 40 (73) | 23 (42) | 49 (89) | 26 (48) | 25 (46) | |
13 (133) | 44 (58) | 32 (43) | 24 (32) | 54 (71) | 27 (35) | 19 (25) | |
3 (30) | 37 (11) | 57 (17) | 7 (2) | 33 (10) | 43 (13) | 23 (7) | |
Relevant medical condition | |||||||
75(883) | 35(310) | 42(243) | 19(205) | 53(466) | 28(243) | 19(170) | |
25(290) | 32(94) | 41(119) | 27(77) | 37(107) | 35(101) | 28(80) | |
Taking medication | |||||||
74 (863) | 38 (324) | 40 (349) | 22 (190) | 53 (454) | 29(249) | 18 (156) | |
26 (310) | 26 (80) | 45 (138) | 30 (92) | 39 (119) | 31 (95) | 31 (94) | |
Previous live birth | |||||||
59 (607) | 30 (184) | 45 (273) | 25 (150) | 50 (300) | 25(154) | 25 (151) | |
41 (428) | 40 (172) | 39 (166) | 21 (90) | 51 (219) | 32(137) | 16 (70) | |
If previous live birth, any conditions/abnormalities | |||||||
64 (227) | 40 (90) | 40 (91) | 20 (46) | 52 (117) | 34 (76) | 14 (31) | |
36 (126) | 39 (49) | 39 (49) | 22 (28) | 46 (58) | 33 (41) | 21 (27) | |
Previous miscarriage, stillbirth or termination | |||||||
70 (654) | 36 (236) | 41 (267) | 23 (151) | 53(345) | 27(175) | 20(131) | |
30 (279) | 29 (81) | 47 (132) | 24 (66) | 43(119) | 32(88) | 25(70) | |
Pregnancy Intention | |||||||
3 (36) | 61 (22) | 11 (4) | 28 (10) | 66(23) | 23(8) | 11(4) | |
24 (272) | 53 (144) | 33 (90) | 14 (38) | 62(166) | 28(76) | 10(27) | |
73 (840) | 27 (230) | 46 (385) | 27 (225) | 45(375) | 30(252) | 25(211) |
*Includes advice on folic acid
**Includes advice on folic acid
Behaviour change | Supplement consumption | |||||
Characteristic %(n) | ||||||
48 (115) | 41 (262) | 31 (346) | 49(572) | 26(301) | 25(285) | |
Age | ||||||
31 (21) | 33 (36) | 24 (67) | 66(187) | 24(68) | 10(29) | |
56 (50) | 41 (113) | 32 (131) | 43(178) | 27(110) | 30(125) | |
57 (37) | 43 (91) | 36 (114) | 45(205) | 27(123) | 28(131) | |
Ethnic group | ||||||
51 (85) | 41 (198) | 32 (219) | 44(306) | 28(196) | 28(194) | |
45 (5) | 36 (10) | 29 (17) | 57(32) | 23(13) | 20(11) | |
44 (7) | 37 (14) | 35 (42) | 55(66) | 20(24) | 25(30) | |
47 (7) | 36 (8) | 23 (23) | 70(69) | 17(17) | 12(12) | |
36 (4) | 40 (12) | 24 (14) | 46(27) | 24(14) | 30(18) | |
Place of birth | ||||||
43 (43) | 40 (106) | 31 (165) | 49(264) | 27(141) | 24(130 | |
52 (63) | 41 (137) | 31 (152) | 47(230) | 25(120) | 28(135) | |
Employment status | ||||||
56 (80) | 41 (192) | 33 (236) | 45(315) | 27(189) | 28(202) | |
15 (4) | 31 (9) | 22 (22) | 63(63) | 23(23) | 14(14) | |
45 (18) | 39 (35) | 27 (51) | 55(110) | 25(50) | 20(39) | |
50 (5) | 45 (10) | 28 (11) | 37(11) | 26(8) | 37(11) | |
Highest educational achievement | ||||||
58 (69) | 43 (201) | 33 (226) | 39(266) | 29(295) | 32(214) | |
44 (22) | 32 (24) | 29 (53) | 62(111) | 18(33) | 20(37) | |
35 (14) | 30 (17) | 28 (35) | 70(90) | 20(26) | 10(12) | |
17 (1) | 0 (0) | 10 (3) | 63(19) | 30(9) | 7(2) | |
Relevant medical condition | ||||||
48(81) | 40(193) | 29(246) | 49(500) | 26(267) | 25(264) | |
48(34) | 42(69) | 35(100) | 57(72) | 27(34) | 16(21) | |
Taking medication | ||||||
48 (87) | 40 (175) | 30 (247) | 53(450) | 26(217) | 21(183) | |
47 (28) | 43 (87) | 32 (99) | 40(122) | 27(84) | 33(102) | |
Previous live birth | ||||||
55 (76) | 45 (176) | 33 (196) | 48(281) | 26(159) | 26(159) | |
37 (29) | 29 (61) | 28 (116) | 51(216) | 25(106) | 24(103) | |
If previous live birth, any conditions/abnormalities | ||||||
40 (17) | 32 (39) | 29 (64) | 45(99) | 27(60) | 28(63) | |
38 (10) | 23 (12) | 28 (35) | 58(73) | 24(30) | 18(22) | |
Previous miscarriage, stillbirth or termination due to abnormalities | ||||||
54 (67) | 38 (154) | 29 (187) | 50(322) | 26(166) | 24(158) | |
41 (32) | 49 (73) | 37 (101) | 41(112) | 27(76) | 32(88) | |
Pregnancy Intention | ||||||
10 (1) | 13 (1) | 0 (0) | 94(33) | 3(1) | 3(1) | |
17 (14) | 12 (15) | 12 (31) | 81(212) | 12(32) | 7(19) | |
68 (99) | 49 (245) | 38 (314) | 37(312) | 31(259) | 32(263) |
Smoking before pregnancy was reported by 21% of women (5% passive only, 10% smoking 0–10 cigarettes per day and 7% smoking more than 10 per day) but nearly half of them (48%,
Overall, 27% of women reported visiting a health professional for advice about getting pregnant although their source of information on a range of preconception health behaviours was more commonly outside the health profession (
Supplements | GP | Health Professional | Family/Friends | Other |
28.6(336) | 6.1(72) | 24.4(286) | 17.2(202) | |
6.4(75) | 3.2(37) | 10.7(125) | 11.8(138) | |
1.1(13) | 0.3(3) | 2.0(24) | 2.8(33) | |
3.8(45) | 1.8(21) | 2.0(23) | 3.1(36) | |
3.4(40) | 0.9(11) | 2.4(28) | 3.3(39) | |
1.4(17) | 0.9(11) | 4.1(48) | 4.4(52) | |
2.0(23) | 0.7(8) | 2.4(28) | 3.8(44) | |
0.5(6) | 0.7(8) | 0.9(11) | 2(23) | |
Other preconception health behaviours | ||||
16.1(189) | 5.2(61) | 22.5(264) | 24.8(291) | |
10.0(117) | 3.9(46) | 5.1(60) | 15.0(176) | |
8.7(102) | 4.4(52) | 13(152) | 22.2(260) | |
13.3(156) | 5.0(59) | 15.3(180) | 26.3(309) | |
13.1(154) | 4.5(53) | 13.6(160) | 23.5(272) | |
6.9(81) | 3(35) | 7.6(89) | 16.4(192) | |
6.9(81) | 2.7(32) | 3.6(42) | 9.0(106) | |
9.4(110) | 2.9(34) | 8.2(96) | 13.7(161) | |
10.0(117) | 5.0(59) | 6.7(79) | 12.4(146) | |
5.8(68) | 2.0(23) | 5.5(64) | 9.5(111) | |
7.6(89) | 3.2(38) | 3.6(42) | 10.4(122) |
*Includes finding out themselves, internet and books etc.
One quarter of the sample reported a relevant medical condition: diabetes, epilepsy, hypertension, HIV, kidney disease, lupus, thyroid disease, sickle cell, obesity, acne, asthma, depression, rheumatoid arthritis or lung disease and/or a treatment with potential teratogenic effects (
Level of health professional input | ||||
Outcome | None | Includes advice on folic acid |
Includes advice on folic acid |
|
30 (95) | 44 (79) | 62 (88) | ||
1 - | 1.86 (1.27–2.72) | 3.77 (2.50–5.71) | ||
1 - | 1.32 (0.77–2.24) | 1.45 (0.84–2.52) | ||
37 (43) | 55 (37) | 63 (35) | ||
1 - | 2.09 (1.14–3.86) | 2.83 (1.46–5.47) | ||
1 - | 1.83 (0.67–5.04) | 0.89 (0.31–2.51) | ||
20 (112) | 37 (123) | 46 (110) | ||
1 - | 2.33 (1.72–3.16) | 3.41 (2.45–4.73) | ||
1 - | 2.42 (1.62–3.61) | 2.18 (1.42–3.36) | ||
37 (209) | 55 (189) | 76 (188) | ||
1 - | 2.09 (1.59–2.75) | 5.31 (3.79–7.44) | ||
1 - | 1.65 (1.12–2.44) | 3.86 (2.35–6.35) |
Adjusted for age, ethnicity, education, previous live birth, medications, previous miscarriage/stillbirth, and LMUP score.
We interviewed 21 health professionals by telephone: four consultants in obstetrics and gynaecology, eight community based consultants (or clinical leads) in sexual and reproductive health, seven general practitioners, one sexual health specialist nurse and one midwife. The key findings related to awareness of guidelines about preconception health and care, responsibility for providing pre-pregnancy care, and existing barriers to provision.
Most of the health professionals we interviewed were able, when asked, to define pre-pregnancy care as counseling women on factors such as exercise, smoking cessation, and reducing alcohol to encourage a healthy pregnancy. One interviewee responded with an interesting, but significant, misinterpretation:
Most interviewees expressed only a vague awareness of guidelines on pre-pregnancy health and care, although a number were able to identify potential sources of information, such as the Department of Health, RCOG, and NICE websites. Some interviewees expressed the wish for guidelines specific to preconception care that would lead to more uniformity in their practice.
While most respondents felt that preconception care was important for optimising pregnancy and birth outcomes, there was a general sense that it was someone else's responsibility (see Interviews 1). Only two interviewees, both clinical leads from sexual and reproductive health, thought that preconception care was definitely within their role and should be part of routine contraceptive care, e.g.:
Interviews 1: Professional responsibility for pre-pregnancy care
Most interviewees felt there should be a targeted public health campaign to promote pre-pregnancy health, and there was a view that women should be encouraged to take the initiative themselves to obtain information about preconception health (see Interviews 2). Specialist preconception services were seldom regarded as the solution.
Interviews 2: Information and Education
Barriers to providing preconception care included women having unplanned pregnancies, lack of knowledge and interest among health professionals and constrained resources, especially in general practice. Raising awareness in people of reproductive age, improving knowledge and confidence through training of health professionals and providing financial incentives for delivery of preconception care in general practice were seen as necessary to improve current provision.
The USA Centers for Disease Control defines preconception care as ‘a set of interventions that aim to identify and modify biomedical, behavioural and social risks to a women's health or pregnancy outcome through prevention and management’
In this study, awareness of preconception health issues was generally low among women and health professionals. The high level of pregnancy planning contrasts with low levels of information acquired about pre-pregnancy health and low uptake of folate, even in women with a poor obstetric history or relevant medical condition. However, we found that the three months before pregnancy was a time when women who smoked cigarettes or drank alcohol were quite likely to cut down or quit these risk behaviours. Furthermore, women who received advice from a health professional before pregnancy were more likely than other women to adopt positive behaviour change
The strengths of this study are the combination of qualitative and quantitative data, the high response rate and collection of data before the outcome of the pregnancy was known. The high response rate may reflect the face-to-face recruitment and interest in the topic, or perhaps long waiting times when attending the antenatal service. We also used a more robust measure of pregnancy planning than most other studies. The London Measure of Unplanned Pregnancy
Weaknesses include retrospective reporting of pre-pregnancy behaviours with the potential for social desirability bias (e.g. over-reporting quitting smoking). The significant association between health professional input and preconception behaviour change could be explained by reporting bias and/or confounding, that is, if women who receive input from health professionals are more likely to report and/or adopt health pre-pregnancy behaviours irrespective of any input received. However, the ‘dose effect’ of health professional advice on changing to a healthier diet and taking folic acid that remained after adjusting for confounding factors (age, ethnicity, education, previous live birth, medications, previous miscarriage/stillbirth, and LMUP score) suggests that input from health professionals can have a positive and independent impact on pre-pregnancy behaviours. A randomised trial of pre-pregnancy advice from health professionals would provide stronger evidence for or against this interpretation.
Much of the literature on preconception health and care derives from research
Studies of preconception health are seldom carried out before conception because of the difficulty of identifying women who are planning a pregnancy and likely to become pregnant within a reasonable time frame. The UK Southampton Women's Study, which recruited 12,445 non-pregnant women aged 20–34, is the only study to have followed women to pregnancy, if it occurred
The high level of pregnancy planning in our study is in keeping with other data from the UK when LMUP scores are compared across similar age groups. Use of the LMUP in a number of studies shows that around two thirds of pregnancies leading to births in the UK are planned
Other studies have reported low awareness among women or reproductive age about folic acid and the conditions (neural tube defects) that it can prevent
A recent study of barriers to the implementation of preconception care by general practitioners in Australia
Awareness of preconception health issues, pregnancy planning and uptake of interventions before pregnancy care are related but distinct issues. All three are required to improve preconception health and pregnancy outcomes. In our study, women who received health professional input did not have greater educational attainment than women with no health professional input; they were more likely to have a relevant medical condition, and more likely to adopt positive behaviour changes. Together these findings suggest that focusing on pre-pregnancy intervention by health professionals would not merely benefit the ‘worried well’ or women with specific medical disorders, but could be an effective approach to addressing important health inequalities relating to smoking, alcohol and other risk behaviours. To strengthen the evidence base for preconception health, future research should focus on evaluating how to implement preconception care effectively and the impact that has on pregnancy and birth outcomes.
COREQ checklist.
(DOCX)
We thank all the women and health professionals who took part in the study. We thank the project team and steering group, Emma Sydenham (
We are also grateful for the support of the Margaret Pyke Trust for their on-going contribution to the programme of reproductive health research at UCL.