INTRODUCTION
Preconception care (PCC) refers to interventions delivered before conception that modify preconception risk factors and reduce the burden of adverse pregnancy outcomes such as low birth weight, spontaneous abortion, and preterm birth.1,2 These interventions may take the form of preconception counselling or education, dietary modification, and supplementary medication during the preconception period.1–4 Previous systematic reviews have shown that PCC interventions provided in hospital and community settings improve pregnancy outcomes4–6 and health knowledge,7 and reduce preconception risk factors. However, less is known regarding the effectiveness of primary care-based PCC.8
As the first point of healthcare contact, primary care providers are ideally placed to provide PCC; however, the effectiveness of primary care-based PCC interventions is unclear.9–11 PCC is often a low priority and not routine practice in primary care in many countries1,8,9,12 and almost all primary care-based PCC interventions are directed towards women.8–10,12–18
As modifiable risk factors including smoking and alcohol consumption may also have an impact on men’s reproductive health19 and sperm quality,20 PCC directed towards reproductive-aged males may also improve pregnancy outcomes. Since the previous review investigating the effectiveness of primary care-based PCC interventions,8 a number of studies evaluating the effectiveness of PCC interventions in primary care have been published. In the current study therefore a systematic review was conducted to evaluate the effectiveness of primary care-based PCC interventions delivered to reproductive-aged females and/or males to improve health knowledge, reduce preconception risk factors, and improve pregnancy outcomes. This builds on the previous review published in 2016 that was limited to females, and which included randomised controlled trials (RCTs) published between July 1999 and July 2015.8
RESULTS
Out of 4622 articles, 1684 duplicates were removed and 134 full-text articles were evaluated for eligibility after title and abstract screening. After full-text screening, 28 articles were included, reporting on 22 RCTs (Figure 1). No additional articles were identified through manually screening reference lists of included articles. The included articles were from the US (n = 8),26–33 Iran (n = 8),34–41 Vietnam (n = 5),42–46 China (n = 2),47,48 the Netherlands (n = 2),49,50 India (n = 1),51 Australia (n = 1),52 and Sweden (n = 1).53 Studies recruited females who were: aged 18–35,40 aged 18–45,26,30 able to conceive,28,33,37 planning to have children,41,47,48,51 planning pregnancy within 9 months,27 a year,31,35,36,38,39,43,44,46 2 years,29 1–5 years,49,50 or after the delivery of their first baby,32,42,45,52 and one study recruited males.53 The majority recruited females planning for pregnancy (n = 16).27,29,31,35,36,38,39,41,43,44,46–51 The type of intervention(s), provider, and outcome(s) measured varied across studies. In some studies the providers were healthcare professionals30,32,43,44,46–49,50–53 whereas in other studies non-healthcare professionals26–29,33–42,45 were trained to deliver the intervention (the latter referred to hereafter as ‘trained facilitators’).
Figure 1. PRISMA flow diagram for papers included in the review. RCT = randomised controlled trial.
The PCC interventions were categorised into brief education (a single education session),26,35–37,49,50,52,53 intensive education (multiple education sessions),27–29,33,34,38–41 supplementary medication,43,44,46 dietary modification,42,45,48,51 and multiple interventions.30–32,47 Studies reported on the effectiveness of the intervention on pregnancy outcomes,29,42,44–46,48,50,51,52 health knowledge,37,38,50,53 and/or preconception risk factors.26–29,30–36,40,47,50 Study characteristics and quality are summarised in Supplementary Tables S2 and S3. According to the ROB 2.025 tool, only two of the RCTs were of high quality (n = 2, 9%), eight were of moderate quality (n = 8, 36%), and the majority were of low quality (n = 12, 55%).
Effectiveness of brief education
Eight articles reported on the effectiveness of brief education (a single session).26,35–37,49,50,52,53 In four of these, healthcare professionals, including GPs,49,50 nurse–midwives,53 and midwives, directed the sessions.52 The other four were directed by non-healthcare professionals, including trained facilitators26,35,36 or the researcher.37 Healthcare professional-directed brief education improved pregnancy outcomes in one of two studies investigating this,50,52 improved health knowledge, and reduced preconception risk factors in females49,50 and males.53
In the one study that reported improved pregnancy outcomes, GP-directed counselling about preconception risk factors decreased adverse pregnancy outcomes including miscarriage, extra-uterine pregnancy, perinatal death, preterm birth, low birth weight, and congenital abnormalities.50 This study also reported increased self-reported folate intake (OR 4.93, 95% CI = 2.81 to 8.66)50 and reduced alcohol use during the first trimester (OR 1.79, 95% CI = 1.08 to 2.97), and self-reported maternal anxiety (OR not reported).49 In the second study, midwife-directed counselling during a home visit did not affect the incidence of preterm birth or low birth weight during subsequent pregnancies.52
Among the 22 RCTs, only one study involved males.53 In this RCT, reproductive life plan-based (RLP) counselling by nurse– midwives increased men’s self-reported awareness of preconception lifestyle risk factors 3 months post-intervention (P<0.0001).53 The RLP tool provides individuals with information concerning reproductive health. After RLP counselling 76% of males in the intervention group reported a positive experience of the counselling, and 77% had received new information.
Non-healthcare professional-directed brief education reduced preconception risk factors in females in the four studies investigating this.26,35–37 Three studies reported on the effectiveness of trained facilitator-directed group workshops on preconception health and folate-focused education.26,35,36 Of these, group workshops in one study had a positive impact on healthy behaviours such as increasing physical activity (P<0.01)36 and the other two reported increased self-reported intake of folate-rich food.26,35 Likewise, a researcher-directed 5–10 min RLP counselling session increased self-reported knowledge of folate intake (P<0.001).37
Effectiveness of intensive education
Nine articles reported on the effectiveness of intensive education.27–29,33,34,38–41 In all of these, sessions were directed by non-healthcare professionals, including trained facilitators27–29,33,34,38–41 and the researcher.41
Non-healthcare professional-directed intensive education improved health knowledge,38 reduced spontaneous pregnancy loss,29 or reduced preconception risk factors in all studies investigating this.27–29,33,34,38–41 In one study, group counselling sessions on preconception risk factors increased self-reported health knowledge of preconception lifestyle risk factors (mean difference 7.8, 95% CI = 8.7 to 6.9).38 In another, weekly counselling sessions on health responsibility for 6 months, followed by monthly sessions until delivery, reduced spontaneous pregnancy loss (OR 0.39, 95% CI = 0.16 to 0.92) and increased self-reported weight loss before pregnancy (P<0.001).29
A 14-week counselling programme around hazardous drinking reduced alcohol-exposed pregnancies at 3-, 6-, and 9-month follow-ups.27 Similarly, two motivational interviewing sessions that aimed to increase participants’ commitment to change hazardous alcohol use reduced risk of alcohol-exposed pregnancies across 9 months (incidence rate ratio 0.620, 95% CI = 0.511 to 0.757).33
Six 2-hour sessions on preconception risk factors also increased self-reported physical activity (P = 0.019).28 In another study, 6-weekly motivational interviewing sessions increased self-reported moderate (P = 0.01) and vigorous (P = 0.02) physical activity,40 and increased self-reported weight loss post-intervention (mean difference −1.457 kg, 95% CI = 2.061 to 0.852).34 Two studies investigated the short-term effect of preconception risk factor counselling on maternal stress levels.39,41 In one study, trained facilitator-directed counselling reduced self-reported maternal stress 4 and 8 weeks post-intervention.39 In the other, researcher-directed counselling improved stress management 1 month post-intervention.41
Effectiveness of supplementary medication
Three articles from one RCT reported on the effectiveness of supplementary medication delivered by village health workers.43,44,46 Women receiving multiple-micronutrients (multi-micronutrient or iron and folate supplements) or iron were compared with females receiving folate only (control). Preconception multiple micronutrient supplementation did not affect the prevalence of low birth weight or preterm birth,46 but a follow-up analysis reported high prenatal and postpartum maternal ferritin levels in the groups supplemented with multiple-micronutrients or iron and folate and these females gave birth to infants with greater iron stores. However, the clinical significance is unclear, as anaemia prevalence did not differ between groups.44
Furthermore, maternal depressive symptoms were low during pregnancy and early postpartum, and there was no difference between the groups. Although the underlying mechanisms are unclear, among females at risk of depression, maternal depressive symptoms were lower in the first and second trimesters of pregnancy in the groups receiving multiple-micronutrients or iron and folate compared with the control group (P<0.05).43
Effectiveness of dietary modification
Four articles reported on the effectiveness of dietary modification.42,45,48,51 In two of these, healthcare professionals including health workers51 or specialists48 delivered the intervention and improvements in pregnancy outcomes were reported. 48.51 In one study, health worker provision of a snack (made from leafy green vegetables, fruit, and milk), provided from >90 days before pregnancy until delivery of baby, in addition to their usual diet, increased infant birth weight (P = 0.046).51 This may have resulted from higher micronutrients, energy, and/or protein levels in the snack provided to the intervention group when compared with the control group.51 Additionally, specialist provision of a diet comprising at least 100 g of mushrooms daily from preconception to the 20th week of gestation reduced gestational hypertension (P = 0.023), preeclampsia (P = 0.014), gestational weight gain (P = 0.017), excessive gestational weight gain (P = 0.032), and gestational diabetes (P = 0.047).48 In the other two articles, non-healthcare professional trained facilitators42,45 delivered macronutrient supplementation from preconception to term. This increased maternal protein, iron, zinc, folate, vitamin A, and B12; however, it did not affect infant birth weight42 or infant growth up to 24 months of age.45
Effectiveness of multiple interventions
Four studies reported on the effectiveness of multiple interventions including supplementary medication such as folate supplementation along with brief30–32 or intensive47 education. Interventions involved provision of folate education via a 15-min GP computerised session,31 brief counselling by a gynaecologist,30 monthly counselling by village doctors,47 or brief counselling by paediatric clinicians.32 All studies reported improvements in self- reported folate intake30–32,47 and one study reported reduction in self-reported binge drinking and smoking.32
DISCUSSION
Summary
To the authors’ knowledge, this is the first systematic review of primary care-based PCC that includes males and also the first to consider the role of the provider in the delivery of primary care-based PCC. Results from 28 articles reporting on 22 RCTs were included incorporating an additional 17 articles published since the last review.8 Most articles in the current review were of low quality and the type of interventions, populations, providers, and outcomes varied substantially between the different studies.
A number of important findings were identified in this review. First, both brief and intensive education on preconception health improved health knowledge and reduced preconception risk factors for females, suggesting either method could be utilised by primary care providers to deliver PCC education. However, the duration of brief education (that is, 5–10 mins, 1 hour, 1 day), intensive education (that is, undertaken over 6 weeks, 14 weeks, 18 months), and the timing of PCC education delivery (that is, 9 months, a year, 2 years, 1–5 years before conception or after the delivery of first baby) varied considerably between studies, so it is not clear which are the most effective.
Second, dietary modification improved pregnancy outcomes by reducing pre-eclampsia48 and increasing birth weight51 in two studies;48,51 however, the studies were of moderate to low quality and more high-quality evidence is required. Multiple interventions including brief or intensive folate education along with folate supplementation increased self-reported folate intake in all studies investigating this,30–32,47 reiterating that primary care providers should encourage supplementary medication, including folate supplements, and intake of folate-rich foods, to all females during the preconception period.
Third, although findings suggest that brief education improves health knowledge among males, more research is required as this is based on only one study.53 Fourth, although in 10 of the studies the intervention was delivered by healthcare professionals, in the majority of studies (n = 12) the intervention was delivered by non-healthcare professionals. In almost all (n = 11) of these latter studies, improved health knowledge,38 reduced preconception risk factors,26–28,33–36 or reduced spontaneous pregnancy loss were reported.29 These results suggest that primary care-based PCC are effective in improving health knowledge and reducing preconception risk factors; trained facilitators could help improve access to PCC in systems that are already struggling to provide care. Finally, although nine studies were found that reported on pregnancy outcomes, only four reported improvements.29,44,48,50 It is unclear whether this is related to the strength of the intervention being delivered or the intervention itself, therefore more evidence is required to understand the effectiveness of primary care-based PCC on improving pregnancy outcomes.
Strengths and limitations
This review was not restricted to a particular region or country, and therefore provides a broad international perspective on the effectiveness of primary care-based PCC interventions. Five databases were systematically searched for literature; however, relevant articles may still have been missed because of the search strategies employed. By limiting the eligibility criteria to RCTs, non-RCTs investigating primary care-based PCC were excluded that may have reported improved pregnancy outcomes.
Comparison with existing literature
This review included 17 additional articles published since the review published in 20168 and 24 additional articles since the last Cochrane review published in 2009.18 Similar to another review of PCC in community settings,4 the current review found that primary care-based PCC interventions are effective in improving health knowledge, increasing here folate intake, and reducing alcohol consumption.
Implications for research and practice
Given the effectiveness of PCC education delivered in primary care at reducing risk factors, brief or intensive PCC education should be mainstreamed for reproductive-aged females and males in primary care.
None of the RCTs in this review targeted PCC for females at high risk of poor pregnancy outcomes based on pre-existing54 medical and lifestyle health indicators. Also there is a lack of understanding about how high-risk females can be systematically identified in primary care. Research investigating how to best identify these females in primary care is therefore warranted.
Lastly, in all studies providing intensive PCC education, interventions were delivered by trained facilitators or researchers. Future studies could explore the role and potential impact of primary care providers, including GPs, nurses and midwives, and trained non-healthcare professionals delivering intensive PCC education.
In conclusion, primary care-based PCC including brief/intensive education, supplementary medication, and/or dietary modification improved health knowledge and reduced preconception risk factors among females, irrespective of the provider. Brief education may also improve health knowledge in males, although more research is needed.
Given the limited number of studies reporting on pregnancy outcomes, further research is required to determine whether primary care-based PCC is effective in improving pregnancy outcomes.