public health Journal of Physiotherapy 64 (2018) 24–32 Research Attitudes, barriers and enablers to physical activity in pregnant women: a systematic rev

public health Journal of Physiotherapy 64 (2018) 24–32

Attitudes, barriers and enablers to physical activity in pregnant women: a
systematic review

Anne L Harrison a,b, Nicholas F Taylor a,c, Nora Shields a,d, Helena C Frawley e,f

a School of Allied Health, La Trobe University; b Physiotherapy Department, Werribee Mercy Hospital; c Allied Health Clinical Research Office, Eastern Health; d Northern Health;
e Department of Physiotherapy, Faculty of Medicine, Nursing and Health Sciences, Monash University; f Centre for Allied Health Research and Education, Cabrini Health,

Melbourne, Australia


Diabetes gestational
Systematic review


Question: What are the attitudes, barriers and enablers to physical activity perceived by pregnant
women? Design: In a systematic literature review, eight electronic databases were searched: AMED,
CINAHL, Embase, Joanna Briggs Institute, Medline, PsycInfo, SPORTDiscus (from database inception until
June 2016) and PubMed (from 2011 until June 2016). Quantitative data expressed as proportions were
meta-analysed. Data collected using Likert scales were synthesised descriptively. Qualitative data were
analysed thematically using an inductive approach and content analysis. Findings were categorised as
intrapersonal, interpersonal or environmental, based on a social-ecological framework. Participants:
Pregnant women. Intervention: Not applicable. Outcome measures: Attitudes and perceived barriers
and enablers to physical activity during pregnancy. Results: Forty-nine articles reporting data from
47 studies (7655 participants) were included. Data were collected using questionnaires, interviews and
focus groups. Meta-analyses of proportions showed that pregnant women had positive attitudes towards
physical activity, identifying it as important (0.80, 95% CI 0.52 to 0.98), beneficial (0.71, 95% CI 0.58 to
0.83) and safe (0.86, 95% CI 0.79 to 0.92). This was supported by themes emerging in 15 qualitative studies
that reported on attitudes (important, 12 studies; beneficial,10 studies). Barriers to physical activity were
predominantly intrapersonal such as fatigue, lack of time and pregnancy discomforts. Frequent enablers
included maternal and foetal health benefits (intrapersonal), social support (interpersonal) and
pregnancy-specific programs. Few environmental factors were identified. Little information was
available about attitudes, barriers and enablers of physical activity for pregnant women with gestational
diabetes mellitus who are at risk from inactivity. Conclusion: Intrapersonal themes were the most
frequently reported barriers and enablers to physical activity during pregnancy. Social support also
played an enabling role. Person-centred strategies using behaviour change techniques should be used to
address intrapersonal and social factors to translate pregnant women’s positive attitudes into increased
physical activity participation. Registration: PROSPERO CRD42016037643. [Harrison AL, Taylor NF,
Shields N, Frawley HC (2018) Attitudes, barriers and enablers to physical activity in pregnant women:
a systematic review. Journal of Physiotherapy 64: 24–32]
© 2017 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article

under the CC BY-NC-ND license (

J o u r n a l o f

jou r nal h o mep age: w ww.els evier .co m/lo c ate/jp hys

Physical activity has substantial benefits for women with
uncomplicated pregnancies, minimal risks, and is recommended in
pregnancy guidelines.1–3 The benefits of physical activity during
pregnancy include improved physical fitness,3–5 reduced risk of
excessive weight gain,6 reduced risk of pre-eclampsia and pre-term
birth,7 reduced low back pain,8,9 improved sleep,10 reduced anxiety
and depressive symptoms,11,12 and improved health perception13

and self-reported body image.14

Physical activity is also important for pregnant women with
comorbidities and complications such as obesity1 or gestational
diabetes mellitus (GDM).15–17 Physical activity assists with weight
control and reduces the risk of GDM in obese pregnant women.1 In
women diagnosed with GDM (a common pregnancy-related
complication occurring in 3.5 to 12% of pregnancies),15,16 physical
1836-9553/© 2017 Australian Physiotherapy Association. Published by Elsevier B.V. This
activity is beneficial as an adjunctive intervention in the
management of glycaemic control.15,17–20 Managing glycaemic
control is critical for reducing adverse effects associated with
poorly controlled GDM.21 Consequently, aerobic exercise per-
formed at moderate intensity for 30 minutes on most days of the
week is recommended for healthy pregnant women,1,3 those with
GDM15,22,23 and those who are overweight or obese.24

Despite well-documented health benefits,1,3–17,24–27 60 to 80%
of pregnant women28–31 – including those who are overweight or
obese31 – and more than 60% of women with GDM32 do not
participate in physical activity as recommended. Pregnant women
from backgrounds other than Caucasian are also less likely to
engage in physical activity.29 However, to improve pregnant
women’s participation in physical activity (ie, leisure time physical
activities or structured exercise programs), we need to understand
their attitudes to it, the reasons why they do not engage in physical
is an open access article under the CC BY-NC-ND license (http://creativecommons.

Box 1. Inclusion criteria.

� Qualitative or quantitative studies
� Full-text article published in a peer-reviewed journal
� Pregnant women whose pregnancy was not high risk a

Outcome measures
� Pregnant women’s attitudes to physical activity b during

� Pregnant women’s perceived barriers and enablers to
physical activity during pregnancy

a High-risk pregnancy was defined as premature labour, in-
competent cervix, persistent bleeding, ruptured membranes,
growth retardation, pre-eclampsia, severe anaemia, placenta
previa after 26 weeks gestation, haemodynamically significant
heart disease or restrictive lung disease.1,2

b Physical activity was defined as leisure time physical activi-
ties and structured exercise programs.

Research 25
activity, and enablers that could be harnessed to design effective
physical activity interventions or programs that facilitate behav-
iour change and thereby improve their participation in physical
activity during pregnancy.

The inclusion of behaviour change techniques into physical
activity interventions has been reported as helpful in improving
physical activity levels during pregnancy.33 Behaviour change
techniques such as goal setting, planning and education to shape
knowledge appear most effective when delivered with face-to-face
feedback about goal achievement.33 However, to facilitate uptake
of these effective physical activity interventions, clinicians need to
know which barriers, enablers and attitudes are common among
pregnant women, so they can effectively target their education and
evidence-based behaviour change strategies. A systematic review
of barriers, enablers and attitudes of pregnant women to physical
activity would provide valuable information to enable clinicians to
effect a positive behaviour change of increased physical activity in
this group.

Identification of women’s attitudes and perceptions of barriers
and enablers to physical activity in pregnancy could be informed by
quantitative or qualitative research approaches. A review that
collates data from studies using either method would benefit from
the advantages of each: improving generalisability and providing
deeper insights into pregnant women’s beliefs and perceptions
about physical activity during pregnancy. Inclusion of qualitative
findings may assist in better understanding the factors that can
influence women’s attitudes and perceptions. Such deeper
understanding would provide valuable insight that clinicians
can use to plan strategies to encourage pregnant women – in
particular at-risk groups of women such as those with GDM – to
participate in physical activity. It would also inform the design of
realistic and acceptable interventions to be tested in an effective-
ness study. No systematic review has collated quantitative data or
provided a meta-summary of attitudes and perceptions of barriers
and enablers to physical activity in pregnant women.

Therefore, the research question for this review was:

What are the attitudes, barriers and enablers to physical activity
perceived by pregnant women, including women diagnosed
with gestational diabetes mellitus?


The review was reported in accordance with the Preferred
Reporting Items for Systematic Reviews and Meta-Analyses
(PRISMA) guidelines,34 the Enhancing Transparency in Reporting
the Synthesis of Qualitative Research (ENTREQ),35 and guided by
information from the Cochrane Qualitative and Implementation
Methods Group.36

Identification and selection of studies

One reviewer (AH) searched eight electronic databases: AMED,
CINAHL, Embase, Joanna Briggs Institute, Medline, PsycInfo, and
SPORTDiscus from database inception until June 2016; and
PubMed from 2011 until June 2016. The search strategy comprised
three key concepts: attitudes, barriers and facilitators/enablers;
physical activity; and pregnancy. For each concept, key words and
MeSH terms were combined with the ‘OR’ operator and the results
were combined with the ‘AND’ operator (see Appendix 1 on the
eAddenda). No limits were applied to the search. Reference lists
from included studies were manually searched for additional
relevant articles. Using Google Scholar and Web of Science, citation
tracking was performed on the included articles to identify any
other relevant articles.

Two reviewers (AH and HF/NS/NT) independently reviewed the
title and abstracts of articles yielded according to the inclusion
criteria presented in Box 1. If eligibility was unclear based on the
title and abstract, a full-text version was obtained and reviewed by
two reviewers independently. Disagreements were resolved by
discussion between reviewers.

Studies using qualitative or quantitative methods were includ-
ed. This integrated approach was used to enable thorough
exploration of the women’s perceptions, given the potential for
qualitative data to complement and add greater meaning to
quantitative findings.36 This was intended to maximise the value of
the findings for those designing interventions to promote physical
activity in pregnant women.37

Assessment of characteristics of studies

Adapted from the McMaster Critical Review Forms for

qualitative and quantitative research,38,39 which include guide-
lines for interpreting the criteria40,41 to facilitate inter-rater
reliability,42 the rating method for key criteria for quantitative
and qualitative studies developed by Imms43 was used to assess
validity and rigor of included studies (Table 1 on the eAddenda).
This form has been used previously in a study exploring similar
phenomena in a different cohort.44 Quantitative studies were rated
on sample, measure and analysis. Qualitative studies were rated on
credibility, transferability, dependability and confirmability, con-
sistent with the criteria for trustworthiness.45 A rating of one (no
evidence of study meeting criterion), two (some evidence or
unclear reporting) or three (evidence of study meeting criterion)
was used to rate each criterion.

All included studies were assessed by two reviewers indepen-
dently (AH and HF/NT) and any disagreements resolved by
discussion until an agreement was reached. Where agreement
could not be reached the findings were discussed with a third
reviewer (NS). In appreciation that studies rated as lower
methodological quality on rating scales can still provide useful
insights based on the data,36 all studies were included regardless of
assessment of methodological quality but study quality was taken
into account in interpretation of the results.

Data were extracted from each study regarding sample size, age,

body mass index, ethnicity, education, gestation, parity, comor-
bidities (GDM, obesity) and physical activity level, where available.
See Table 2 on the eAddenda.

Data extraction and analysis

Data were extracted from the included articles using a
standardised form. Data were extracted by one reviewer (AH),

Harrison et al: Pregnant women’s attitudes to physical activity26
summarised into tables and independently checked by a second
reviewer (HF/NT). Qualitative and quantitative data were analysed

Analysis of qualitative data
Qualitative data on attitudes, barriers and enablers were

synthesised using an inductive approach and synthesised into
themes and sub-themes providing a meta-summary. An inductive
approach provides a systematic process for analysing qualitative
data, thereby deriving and summarising findings that are reliable,
valid and linked to the research objectives.46

In preparation for analysis, two reviewers (AH and NS)
independently read, re-read, reviewed and made notes to
familiarise themselves with the content and the context from
which the data arose. The data were transcribed verbatim into an
electronic spreadsheet. Following this, the two reviewers inde-
pendently derived initial coding categories, based on emerging
themes. This coding was derived directly from words, phrases or
paragraphs, as the primary aim was to identify the expression of
attitudes and perceptions consistent with the review objectives. To
facilitate consistency of coding, a ‘code-book’ of code names based
on emerging themes and accompanying definitions to guide
consistent interpretation was developed. To enhance the trust-
worthiness of the analysis, an audit trail was kept and an iterative
process was followed involving: independently coding data;
comparing inter-coder agreement; discussing and refining the
coding scheme; and augmenting with interpretive memos. This
iterative process was continued until sufficient coding consistency
and agreement were achieved. Following this, the agreed coding
rules were applied to all of the data by one reviewer/coder (AH)
and independently checked by a second reviewer/coder (NS).

The themes were grouped in three categories: intrapersonal
(eg, physical, psychological), interpersonal (eg, influences from
family, friends, health professionals, social and cultural norms) and
environmental (eg, access to facilities, built environment, policy
and program such as cost), based on a social-ecological model.47,48

An inductive approach was used to categorise the data into themes
and sub-themes under this framework. Data were included under
more than one theme if it was considered that the data satisfied the
definition of more than one theme. For example, ‘participants
considered physical activity important for self and baby’49 with an
accompanying description of benefits was included under the
Records identified through
database searching (n = 3045)


Records after duplicates remov

Records screened (n = 1880 )


Potentially relevant articles
retrieved for full-text evaluation
(n = 103)


Studies included in synthesis
(n = 49 articles)
(n = 47 studies)

Figure 1. PRISMA flow diagram showing id
themes of ‘important’ and ‘beneficial’. Once all data were analysed,
a count for each theme was conducted, checked and recorded.

Analysis of quantitative data
As the majority of quantitative studies reported data expressed

as percentages, these data were synthesised by meta-analyses of
proportions using a random-effects model to account for
heterogeneity. Statistical heterogeneity in each meta-analysis
was reported using the I2 statistic with values > 50% considered
indicative of statistical heterogeneity. The quantitative data were
grouped under the categories: intrapersonal, interpersonal and
environmental, consistent with qualitative analysis. Data collected
using Likert scales were synthesised descriptively.


Flow of studies through the review

The search strategy yielded 3045 articles, including papers in
languages other than English. After screening of titles and
abstracts, 99 full-text articles were retrieved and following
reference checking and citation tracking, four additional articles
were identified totalling 103 articles for full-text review. After
review of these 103 full texts, 54 articles were excluded. Following
this process, 49 articles presenting the results of 47 discrete studies
were included in the review (Figure 1).49–97

Characteristics of included studies

Twenty-two articles reporting data from 21 discrete studies

used qualitative methods, and seven studies used mixed meth-
ods.60,61,66,75,83,91,97 Three49,53,68 of these 28 studies provided
evidence to satisfy all four quality criteria for qualitative studies.
Six studies (reported in seven articles)52,61,66,82,85,91,92 satisfied
three criteria with some evidence of meeting the fourth (see
Table 2 on the eAddenda). These studies reported evidence of
prolonged engagement, a variety of data collection methods,
member checking, detailed descriptions of participants, settings,
processes, analyses, audit trails, reflection, peer review, and
triangulation. All qualitative or mixed-methods studies demon-
strated at least some evidence of trustworthiness.
dditional records identified
hrough other sources (n = 4)

ed (n = 1880 )

ecords excluded (n = 1777 )

ull-text articles excluded (n = 54)
• participants not pregnant at data

collection (n = 23)
• not specifically describing attitudes,

barriers, or enablers (n = 20)
• conference abstract only (n = 9)
• commentary paper (n = 1)
• extra duplicate (n = 1)

entification and selection of studies.34

Table 3
Content analysis summary of qualitative data on attitudes, barriers and enablers to physical activity during pregnancy from 28 studies (reported in 29 articles) that used
qualitative methods.

Attitudes Barriers Enablers

(15 studies) n (27 studies) n (21 studies) n

Important Intrapersonal Intrapersonal
Important/necessary 12 Fatigue 20 Easier labour/delivery 13
As important as diet in pregnancy 1 Safety/fears 20 Maternal health and wellbeing 12
Important for self and baby 1 Pregnancy symptoms/discomforts 19 Weight control 9

Lack of time 17 Ease pregnancy symptoms/discomforts 7
Beneficial Lack of motivation 13 Confidence/physical activity habit 7
Beneficial for women Lack of confidence 8 Baby’s health 6
For healthy pregnancy 10 Lack of knowledge 4 Appearance 5
Fitness and staying in shape 4
For labour/birth 3 Interpersonal (social) Interpersonal (social)
Wellbeing/enjoyment 2 Social support Social support
For pregnancy symptom relief 1 Lack support of family/friends/others 9 Support of partner 10

Beneficial for baby 4 Lack support of partner 3 Support of family/friends/others 9
Lack company 1 Socialisation with other pregnant women 5

Safety Informational Company for walks 1
Need to modify physical activity in pregnancy 2 Lack physical activity information 2 Informational
Walking considered best/safest 1 Conflicting advice 2 Advice from doctor 4

Lack of advice from professionals 2 Unambiguous advice 3
Norms Reassuring advice 3
Social norms 2 Social influence
Cultural influence 1 Socialisation 4

Responsibilities Peer pressure 1
Work commitments 7 Responsibilities
Childcare 5 Fewer commitments, more time 3
Families 2 Childcare support 1

Environmental Environmental
Access Access
Lack access to facilities/resources 11 Access to facilities/resources 9
Lack safe place to be physically active 3 Weather

Weather Good weather 4
Bad weather, hot weather 9 Policy/programs

Policy/programs Pregnancy-specific programs 6
Affordability 7
Lack of pregnancy-specific programs 2

Categories (unindented) contain themes (in italics) and subthemes. n = number of studies reporting each theme or subtheme.

Research 27
Twenty articles reporting data from 19 discrete studies used
quantitative methods. All of these studies and the seven mixed-
methods studies provided some evidence toward meeting at least
one of the three criteria (See Table 2 on the eAddenda).
Four58,64,70,80 studies met all three quality assessment criteria
for quantitative studies and 11 studies met two criteria and
provided some evidence toward the other.50,54,60–62,72,74,76,87,91,97

The characteristics of the participants in the included studies

are detailed in Table 2 (see the eAddenda for Table 2). These studies
included 7655 women representing a range of age groups,
gestational age, parity, body mass index, countries and cultural,
educational and socioeconomic backgrounds. Race/ethnicity and
socio-economic background were broad and women were from
rural, metropolitan, lower socio-economic and more affluent areas,
accessing care in public and private health systems. Six studies
(776 participants), reported in seven articles, studied only
pregnant women who were overweight or obese.56,63,67,87,91,92,95

The range of gestational age reported across studies was from
4 to 41 weeks gestation, providing good representation of women
from across the three trimesters of pregnancy. From the 35 studies
reporting on parity, an average of 55% of participants were
expecting their first baby. Although studies may have potentially
included women with GDM, four studies (77 participants)
explicitly included only pregnant women diagnosed with GDM,
and measured and reported findings for this specific group of
pregnant women.51,55,57,72

Attitudes to physical activity in pregnancy

Attitudes to physical activity in pregnancy were reported in
29 studies (5275 participants): 13 qualitative, 14 quantitative and
two mixed-methods studies. The only attitudes reported in the
13 qualitative studies were that physical activity in pregnancy is
important, beneficial and safe (Table 3). The majority of
participants reported a positive attitude to physical activity in
pregnancy. Based on pooling of the proportion data from
11 quantitative studies, most women identified physical activity
as important (0.80, 95% CI 0.52 to 0.98), beneficial (0.71, 95% CI
0.58 to 0.83) and safe (0.86, 95% CI 0.79 to 0.92), as presented in
Figure 2. (The numerical data used to generate Figure 2 are
available in Table 4 on the eAddenda. The individual meta-analyses
of proportions for each attitude are available in Appendix 2 on the
e-Addenda.) The meta-analyses had high I2 values with
most > 90%. Five studies used Likert scales to rank attitude, and
all reported a positive attitude to physical activity during
pregnancy.54,70,72,74,87 A positive attitude to the importance and
benefits of physical activity during pregnancy was also consistent
across studies reporting this outcome for overweight and obese
pregnant women (n = 4),63,87,91,95 specific race or ethnic popula-
tions (n = 3)70,77,89 and women with GDM (n = 1).72

Barriers to physical activity in pregnancy
Barriers to physical activity in pregnancy were reported in

41 studies (6771 participants; 20 qualitative, 14 quantitative and
7 mixed methods). The most frequent barriers cited were
intrapersonal: fatigue, lack of time and pregnancy discomforts
such as nausea, pain and awkwardness due to weight gain and
increasing size as pregnancy progressed, and less frequently safety
concerns such as the type and intensity of physical activity that is
considered safe during pregnancy and fears for self, the pregnancy
and the baby (Table 3 and Figure 2). (The numerical data used to
generate Figure 2 are available in Table 4 on the eAddenda. The
individual meta-analyses of proportions for each barrier are
available in Appendix 2 on the e-Addenda.) These same barriers


PA in pregnancy is important

Positive attitude to PA in pregnancy

PA beneficial for pregnant women

PA beneficial for baby
Low to moderate intensity PA is safe


Intrapersonal barriers


Lack of time

Pregnancy discomforts


Lack of motivation

Lack of confidence/PA habit

Lack of knowledge

Interpersonal barriers

Lack support of partner

Lack support of family/friends

Lack information

Childcare responsibilities

Work commitments

Environmental/Policy/Program barriers

Bad or hot weather

Lack access to facilities/affordability


Intrapersonal enablers

Maternal health and wellbeing

Baby’s health

Help with labour

Ease pregnancy discomforts

Weight control


Confidence/habit of PA

Interpersonal enablers

Support of partner

Support of family/friends/company for walks

More information

Fewer commitments/more time

Environmental/Policy/Program enablers

Pregnancy specific programs

Access to facilities

Good weather

0 0.25 0.5 0.75 1





































































Figure 2. Forest plot of estimates of the proportion of pregnant women that report each attitude, barrier or enabler in relation to physical activity during pregnancy. Each
estimate is the result of a ‘meta-analysis of proportions’ including the number of studies shown. Each meta-analysis of proportions is reported in detail in Appendix 2 (see
eAddenda for Appendix 2). PA = physical activity.

Harrison et al: Pregnant women’s attitudes to physical activity28
were also highlighted in four quantitative studies that collected
data using Likert scales.58,62,70,87

These themes also arose consistently across studies including:
participants from particular races or ethnic populations (2371 parti-
cipants);51,70,71,73,77,80,81,89,93overweightand obese pregnant women
(802 participants);56,63,67,87,91,92,95 and women with GDM (77 parti-
cipants).51,55,57,72 (For more detailed data on barriers in women with
GDM, see Tables 5 and 6 on the eAddenda). In addition, for women
from ethnic backgrounds other than Caucasian, safety concerns for
their pregnancy and baby emerged as a theme.51,70,71,73,81,89,93 One
study77 of low-income African-American women reported the
cultural norm of lack of exercise habit and socio-economic factors
of lack of affordable and safe places for physical activity as specific
barriers for them. Lack of safe and affordable places to be physically
active and suitable exercise classes also emerged as a barrier to
physical activity in one study95 that included overweight and obese
pregnant women, while another study67reported lack of confidence,
motivation and knowledge as important barriers. Lack of access to
facilities was identified as a barrier to physical activity in pregnancy
in one study)72 compared to the whole sample (0.06, CI 0.00 to 0.17;
245 of 3222 participants).

Research 29
Fewer sub-themes for barriers emerged in the interpersonal
(social) and environmental or policy/programs categories. Lack of
social support from family or friends, lack of information and work
responsibilities were the most frequently cited interpersonal
barriers. Environmental barriers were lack of access and (unfa-
vourable) weather.

Enablers to physical activity in pregnancy
Enablers to physical activity during pregnancy were reported in

36 studies (5730 participants; 17 qualitative, 15 quantitative and
4 mixed methods). Intrapersonal factors were the most frequently
reported enablers (Table 3 and Figure 2). (The numerical data used
to generate Figure 2 are available in Table 4 on the eAddenda. The
individual meta-analyses of proportions for each enabler are
available in Appendix 2 on the e-Addenda.) The intrapersonal
factors included maternal and foetal health and wellbeing, easing
pregnancy discomforts, and an easier labour and birth. Two
studies62,87 reported data using Likert scales and found maternal
health and wellbeing, decreasing stress, improving fitness and
improving appearance were strong enablers of physical activity.
These findings were consistent with the data reported by pregnant
women who: were overweight or obese;56,63,67,87,91,92,95 were from
particular ethnic or racial groups;71,81,89,93 or had been diagnosed
with GDM.

Interpersonal enablers were often cited. Social support was the
most frequently cited interpersonal enabler of physical activity,
particularly partner support and family/friend’s support. This was
also a specific theme among overweight and obese women,67 and
the predominant theme in two studies (416 participants)51,80 of
particular racial groups and in three studies (72 participants)51,55,72

including women with GDM (for more detailed data on enablers in
women with GDM see Tables 5 and 6 on the eAddenda). In these
studies, interpersonal enablers such as support from others (eg,
walking with a partner) were reported more frequently than
intrapersonal factors.

Six qualitative studies49,53,67,77,82,96 and participants from
quantitative studies81,94 reported …

Looking for this or a Similar Assignment? Click below to Place your Order