Women's autonomy in healthcare decision-making and healthcare seeking behaviour for childhood illness in Ghana: Analysis of data from the 2014 Ghana Demographic and Health Survey.

INTRODUCTION: The capacity of women to decide on their healthcare plays a key role in their health. In this study, we examined the association between women's healthcare decision-making capacity and their healthcare seeking behaviour for childhood illnesses in Ghana.
MATERIALS AND METHODS: We used data from the 2014 Ghana Demographic and Health Survey. A total sample of 2,900 women with children less than 5 years was used for the analysis. Data were processed and analysed using STATA version 14.0. Chi-square test of independence and binary logistic regression were carried out to generate the results. Statistical significance was pegged at 95% confidence intervals (CIs). We relied on the 'Strengthening the Reporting of Observational Studies in Epidemiology' (STROBE) statement in writing the manuscript.
RESULTS: Out of the 2,900 women, approximately 25.7% could take healthcare decisions alone and 89.7% sought healthcare for childhood illnesses. Women who decided alone on personal healthcare had 30% reduced odds of seeking healthcare for childhood illnesses compared to those who did not decide alone [AOR = 0.70, CI = 0.51-0.97]. With age, women aged 45-49 had 69% reduced odds of seeking healthcare for childhood illnesses compared to those aged 25-29 [AOR = 0.31, CI = 0.14-0.70]. Women from the Northern and Upper West regions had 72% [AOR: 0.28, CI: 0.11-0.70] and 77% [AOR: 0.23, CI: 0.09-0.58] reduced odds of seeking healthcare for childhood illnesses respectively, compared to those from the Western region.
CONCLUSION: Ghanaian women with autonomy in healthcare decision-making, those who were older and those from the Northern and Upper West regions were less likely to seek healthcare for childhood illness. To reduce childhood mortalities and morbidities in Ghana, we recommend educating women such as those who take healthcare decisions alone, older women and women from deprived regions like the Northern and Upper West regions on the need to seek healthcare for childhood illnesses.

Introduction

Childhood illness and death have become a worldwide health priority and a Sustainable Development Goal [1,2]. Target two of the Sustainable Development Goal three (SDG 3) aims at ending preventable neonatal and under-five deaths by 2030 [3]. Studies have shown that, mortality and morbidity in children under five years have seen significant progress in the last two decades from 12.7 million in 1990 to 5.9 million in 2015 [4-6]. In most countries in sub-Saharan Africa and Central and South Asia, children are almost 80% more likely to die before the age of five than children in the high income countries [1,7]. In Ghana, the under-five mortality rate is still high with a rate of 60 deaths per 1000 live births in 2014 [8]. In most sub-Saharan African countries, including Ghana, acute childhood illnesses which include acute respiratory infections, diarrhoea, malaria, and meningitis are the leading medical causes of infant and child illnesses and deaths [1]. These conditions are both preventable and treatable with the recommended point of care and treatment at the health facility [8-11].

These illnesses cause rapid and serious physiological derangement on the baby and the time taken to seek supportive management is of essence [1]. It is therefore important to improve access to skilled health professionals, and appropriate health care-seeking behaviour of mothers in order to reduce childhood illnesses and deaths [5,12]. The risk of mortality from childhood illnesses when complicated is high and it has been established that early diagnosis and prompt treatment should occur within 24 hours of the onset of these illnesses [13]. In most low-and middle-income countries, about 10% of children die as a result of infectious diseases before their fifth birthday [14]. The causes of childhood illnesses and deaths can reduce with a timely healthcare seeking behaviour and decision making capacity of women and their families [13].

Literature shows that health seeking behaviour for childhood illnesses and deaths, and women healthcare decision-making capacity have similar associated factors including socio-economic and demographic variables [15-18]. In low-and middle-income countries, women’s healthcare decision-making capacity is very important for advancement in maternal and child health outcomes and women empowerment [19,20]. The societal norms, culture, religion and other socio-cultural indicators coupled with the availability of health facilities often determine the circumstances under which women would have capacity to decide on their healthcare and healthcare behaviour of their sick child [22-24]. In Ghana, studies have shown that despite men’s authority in household decision-making, women are active players in the household decision-making process [25,26]. It has also been found that decision-making autonomy is aligned with holistic wellbeing especially in the aspect of maternal and child health and that healthcare decision-making of women in Ghana plays a role in their healthcare seeking behaviour [27]

Studies have revealed that the ability of women in low-and middle-income countries to seek healthcare services is dependent on the child’s health and some other factors that interact to influence their healthcare decision making capacity and health seeking behaviour for their sick children [28-31]. These factors are not limited to their socio-economic status, perception towards modern healthcare treatment, level of education, parents’ literacy level, size of the family, perceived severity of the illness and previous experience of child illness and death [28,32]. The decision-making power with access to and adequate control over economic resources by women of children have a wide positive impact on health seeking behaviour for childhood illness [33,34]. In the same way, studies from sub-Saharan Africa suggest that children of women with limited decision-making capacity in health care are at higher risk of malnourishment, child mortality and negative paediatric health [21,35,36].

In Ghana, several studies have examined the link between women’s decision-making capacity and the use of maternal and child healthcare services including skilled birth attendance [27] and antenatal care, delivery and postnatal care [26]. Despite the link between healthcare decision-making capacity and health care seeking for childhood illness in other studies outside Ghana and the high under-five mortality rate in Ghana attributed to childhood illnesses, it appears no study in Ghana has explored the link between women’s autonomy in healthcare decision-making and health care seeking for childhood illness. In this study, we examined the association between women’s autonomy in healthcare decision-making and their health care seeking for childhood illnesses in Ghana.

Materials and methods

Data source

Data for the study was obtained from the 2014 Ghana Demographic and Health Survey (GDHS). Specifically, data from the children’s recode file was used for the study. The GDHS is a nationwide survey conducted every five years since it began in 1988. The survey gathers information on fertility, family planning, infant and child mortality, maternal and child health as well as nutrition. The GDHS is designed to provide adequate data to monitor the population and health situation in Ghana. The survey adopts a two-stage sampling design. The first stage is characterised by the selection of clusters (427) across urban and rural (211) locations from the entire nation. These made up the enumeration areas for the study. The second stage involved the selection of households from the predefined clusters, and this resulted in the selection of 12831 households. Out of the 12831 households, a total of 5,884 women with children less than 5 years were sampled. However, in this study, only childbearing women whose children had either diarrhoea/cough/fever in the last 2 weeks prior to the survey were considered and this translated into a total sample of 2,900 women. Hence, childbearing women whose children did not have diarrhoea/cough/fever in the last 2 weeks were excluded from the study.

Study variables

Outcome variable

Healthcare seeking for childhood illness was the outcome variable in this study. This variable was obtained by creating a composite variable which comprised “care seeking for diarrhoea in the last 2 weeks” and “care seeking for cough/fever in the last 2 weeks”. Care-seeking for diarrhoea was derived from two sets of questions under the child immunisation, health and nutrition section of the DHS women’s questionnaire. The first question was “Has child had diarrhoea in the last 2 weeks?” The answers were ‘Yes’ and ‘No’. Mothers who answered “Yes”, were asked the second question–“Did you seek advice or treatment for the diarrhoea from any source?” The answers were ‘Yes’ and ‘No’.

Healthcare seeking for cough/fever was derived in a similar manner to diarrhoea. In the DHS, the first question posed was “Has child had an illness with a cough/fever at any time in the last 2 weeks?” The answers were ‘Yes’ and ‘No’. Mothers who answered ‘Yes’, were further asked “Did you seek advice or treatment for the cough/fever from any source?” The answers were ‘Yes’ and ‘No’ [8].

A composite variable was created and coded as “1” if care was sought in at least one of the two circumstances and “0” if no care was sought in all the two circumstances [37]. This helped to generate the outcome variable (healthcare seeking for childhood illnesses).

Independent variables

The key independent variable for the study was women’s autonomy in healthcare decision-making. This was derived from the variable “decision on personal health care”. To obtain this variable, women were asked in the DHS “who usually decides on respondent’s health care” with five responses. These were “respondent alone”, “respondent and husband/partner”, “husband/partner alone” “someone else” and “other”. This was coded as “not alone = 0” and “alone = 1”. A major limitation of this variable is that it does not provide the opportunity to fully interrogate the types of women and men who tend to agree or disagree on decisions, nor the sources of disagreement [38]. Apart from the key independent variable, ten women characteristics and four child’s characteristics were also considered based on their availability in the dataset. The women characteristics were age [28], region [39], educational level [28,30], marital status [3,40], wealth index [30], parity [41], ethnicity [41], place of residence [42], religion [40] and occupation [42]. The child’s characteristics were birth order of the child, child’s weight (weight of the child at birth) [43-46], child’s twin status and child’s sex [47]. Parity was coded as one birth (1), two births (2), three or more births (3). Education was recoded as no formal education (0), primary (1) and secondary/higher (2). Occupation was recoded as not working (0) and working- managerial, clerical, sales, agriculture, services and manual (1). In addition, religion was recoded into Christianity (1), Islam (2), Other (3). Child’s birth order was recoded into first child (1), second child or more. Child’s weight was recoded into below 2.5kg (1) and 2.5kg and above (2). This weight was generated based on studies that have considered a child’s weight less than 2.5kg as low birth weight and those 2.5kg and above as non-low birth weight [35-38]. Child’s twin status was recoded into single birth (1) and multiple birth (2).

Statistical analyses

Data were processed and analyzed using STATA version 14.0, employing inferential and descriptive statistics. The analysis was done in two steps. First, descriptive statistics (frequency and percentages) were used to describe the characteristics of the respondents and their association with healthcare seeking for childhood illnesses was assessed using chi-square (see Table 1). The second step involved the use of multivariable logistic regression to assess the association between the independent variables and the dependent variable. The selection of the variables into the regression model was not influenced by their statistically significant associations with the outcome variable at the bivariate analysis but on their availability in the dataset and their significant association with healthcare seeking behaviour for childhood illnesses as found in previous studies [3,40,42]. The multivariable logistic regression analysis was conducted using three models. Model I was fitted with the key independent variable (women’s autonomy in healthcare decision-making) and the dependent variable (healthcare seeking for childhood illnesses). In Model II, we adjusted for possible confounders (maternal characteristics) to assess how they affect the relationship between healthcare-seeking and women’s autonomy in decision-making (Table 2). In the third model (Model III), which was the complete model, we adjusted for possible confounders (all the independent variables) to assess how they affect the relationship between healthcare-seeking and women’s autonomy in decision-making. The coefficients of the models were exponentiated to derive adjusted odds ratios (AORs). Statistically significant results were assessed at 95% confidence level. To check for high correlation among the explanatory variables, a test for multicollinearity was carried out using the variance inflation factor (VIF) and the results showed no evidence of high collinearity (Mean VIF = 1.41, Maximum VIF = 2.78, and Minimum VIF = 1.01). Sample weight (v005/1,000,000) was used to correct for over and under-sampling while the SVY command was used to address the complex survey design and generalizability of the findings. We relied on Strengthening the Reporting of Observational Studies in Epidemiology’ (STROBE) statement in writing the manuscript.

Table 1

Sample characteristics and women’s autonomy in healthcare decision-making and healthcare-seeking for childhood illness according to sample characteristics.

Variables	N	%	Healthcare seeking for childhood illnesses		χ2 (P-value)
			No	Yes
Women’s autonomy in healthcare decision-making					8.97(0.003)
Not Alone	2154	72.26	6.77	93.23
Alone	746	25.74	10.31	89.69
Age					13.74(0.033)
15–19	54	1.85	11.29	88.71
20–24	362	12.47	7.20	92.80
25–29	783	27.01	5.52	94.48
30–34	820	28.26	7.83	92.17
35–39	589	20.31	8.64	91.36
40–44	251	8.65	8.37	91.63
45–49	42	1.45	16.67	83.33
Region					32.99 (<0.001)
Western	265	9.14	3.11	96.89
Central	267	9.21	4.06	95.94
Greater Accra	598	20.63	6.23	93.77
Volta	229	7.88	4.35	95.65
Eastern	214	7.36	9.68	90.32
Ashanti	668	23.05	11.70	88.30
Brong Ahafo	252	8.68	8.26	91.74
Northern	189	6.51	9.84	90.16
Upper East	138	4.76	6.59	93.41
Upper West	80	2.77	10.39	89.61
Occupation					0.03(0.866)
Not working	497	17.13	7.37	92.63
Working	2,403	82.87	7.59	92.41
Ethnicity					0.98(0.806)
Akan	1,458	50.29	6.98	93.02
Ga-Adangbe	197	6.78	7.46	92.54
Mole-Dagbani	501	17.26	7.89	92.11
Others	745	25.67	8.07	91.93
Marital status					0.01(0.929)
Married	2,185	75.36	7.57	92.43
Cohabiting	715	24.64	7.47	92.53
Residence					2.67(0.102)
Urban	1,653	57.01	8.33	91.67
Rural	1,247	42.99	6.73	93.27
Educational level					1.28(0.526)
No formal education	599	20.66	7.50	92.50
Primary	466	16.08	8.70	91.30
Secondary/higher	1,835	63.26	7.19	92.81
Wealth quintile					4.30(0.367)
Poorest	424	14.64	8.19	91.81
Poorer	414	14.29	5.42	94.58
Middle	539	18.59	7.43	92.57
Richer	691	23.82	8.48	91.52
Richest	831	28.66	7.73	92.27
Religion					1.15(0.562)
Christianity	2,302	69.02	7.24	92.76
Islam	502	22.20	8.32	91.68
Other	96	4.41	8.87	91.13
Parity					0.63(0.729)
One birth	437	15.07	6.77	93.23
Two births	704	24.29	7.33	92.67
Three births	1,759	60.64	7.83	92.17
Child’s birth order					0.08(0.784)
One	2,184	75.30	7.63	92.37
Two or more	716	24.70	7.31	92.69
Child’s weight					1.03(0.310)
Below 2.5kg	263	9.09	9.09	90.91
2.5kg and above	2,637	90.91	7.39	92.61
Child’s Twin status					0.02(0.899)
Single birth	2,736	94.35	7.57	92.43
Multiple birth	164	5.65	7.28	92.72
Child’s sex					1.69(0.194)
Male	1,534	52.90	8.15	91.85
Female	1,366	47.10	6.88	93.12

Source: Computed from 2014 GDHS

Table 2

Bivariate and multivariable logistic regression analysis results.

Variables	Model I COR 95% CI	Model II AOR 95% CI	Model III AOR 95% CI
Women’s autonomy in healthcare decision-making
Not Alone	1	1	1
Alone	0.63** [0.47–0.86]	0.70* [0.51–0.97]	0.70* [0.51–0.97]
Age
15–19		0.45 [0.19–1.09]	0.45 [0.19–1.08]
20–24		0.73 [0.19–1.09]	0.74 [0.45–1.23]
25–29		1	1
30–34		0.64 [0.42–1.00]	0.65 [0.42–1.00]
35–39		0.59*[0.37–0.95]	0.60*[0.37–0.97]
40–44		0.60 [0.34–1.08]	0.61 [0.34–1.09]
45–49		0.30**[0.14–0.68]	0.31**[0.14–0.70]
Region
Western		1	1
Central		0.79 [0.31–2.01]	0.80 [0.32–2.04]
Greater Accra		0.65 [0.26–1.63]	0.65 [0.26–1.64]
Volta		0.91[0.35–2.38]	0.92 [0.35–2.42]
Eastern		0.35*[0.15–0.83]	0.36*[0.15–0.85]
Ashanti		0.28**[0.13–0.62]	0.29**[0.13–0.63]
Brong Ahafo		0.38* [0.17–0.87]	0.39* [0.17–0.88]
Northern		0.27** [0.11–0.69]	0.28*[0.11–0.70]
Upper East		0.42 [0.17–1.03]	0.42 [0.17–1.03]
Upper West		0.23**[0.09–0.57]	0.23**[0.09–0.58]
Occupation
Not working		1	1
Working		1.04 [0.70–1.56]	1.05 [0.70–1.57]
Ethnicity
Akan		1	1
Ga-Adangbe		0.88 [0.42–1.85]	0.88[0.42–1.86]
Mole-Dagbani		1.02 [0.60–1.75]	1.03[0.60–1.76]
Others		0.73 [0.48–1.11]	0.74 [0.49–1.13]
Marital status
Married		1.11 [0.75–1.65]	1.12[0.75–1.65]
Cohabiting		1	1
Residence
Urban		0.84 [0.55–1.29]	0.84 [0.55–1.28]
Rural		1	1
Educational level
No formal education		1.05 [0.69–1.61]	1.05 [0.68–1.60]
Primary		0.74 [0.50–1.10]	0.74 [0.50–1.11]
Secondary/higher		1	1
Wealth
Poorest		0.78 [0.46–1.34]	0.78 [0.46–1.34]
Poorer		1	1
Middle		0.69 [0.40–1.20]	0.69 [0.40–1.20]
Richer		0.63 [0.3401.17]	0.63 [0.34–1.18]
Richest		0.69 [0.34–1.38]	0.69 [0.34–1.39]
Parity
One birth		1	1
Two births		0.89 [0.54–1.47]	0.88 [0.53–1.47]
Three or more births		0.95 [0.57–1.60]	0.92 [0.54–1.60]
Religion
Christianity		1	1
Islam		1.12 [0.76–1.65]	1.12 [0.76–1.65]
Other		0.74 [0.38–1.45]	0.73 [0.37–1.43]
Child’s Birth order
One			0.98 [0.69–1.39]
Two or more			1
Child’s Weight
Below 2.5 kg			0.81 [0.51–1.29]
2.5 kg and above			1
Child’s Twin status
Single birth			0.84 [0.44–1.62]
Multiple birth			1
Child’s sex
Male			0.83 [0.63–1.10]
Female			1
N	2,900	2,900	2,900
pseudo R2	0.005	0.041	0.043

Exponentiated coefficients; 95% confidence intervals in brackets, COR = Crude Odds Ratio, AOR = Adjusted Odds Ratio, 1 = Reference category,

* p < 0.05,

** p < 0.01,

*** p < 0.001

Model 1: Bivariate; Model 2: Adjust for maternal characteristics; Model 3: Adjust for maternal and child characteristics.

Source: Computed from 2014 GDHS

Ethics approval

The DHS reported that ethical approval was granted by the Institutional Review Board of ICF International and Ethical Review Committee of Ghana Health Service. We further obtained permission from the DHS Program for use of this data for the study.

Results

Sample characteristics

Table 1 shows results on the distribution of healthcare seeking for childhood illness across women’s autonomy in healthcare decision-making and their background characteristics. Majority of the respondents (72.3%) did not decide on their healthcare alone, 28.3% of participants were aged 30–34, 23.1% were in the Ashanti region, 83.9% were working and 50.3% were Akans. More than three-quarter (75.4%) of the participants were married, 57.0% were in urban areas, 63.3% had secondary/higher education, 28.7% were in the richest wealth quintile and 69.0% were Christians. With the child characteristics, 90.9% of the participants had children who weighed 2.5kg and above, 94.4% of the participants had single births and 52.9% of the children were male.

Healthcare-seeking for childhood illness

The results indicate that 93.2% of women who had no independent household decision-making capacity sought for healthcare for childhood illnesses. Women aged 25–29 had the highest prevalence of healthcare seeking for childhood illnesses (94.5%). Women from the Western region had the highest prevalence of healthcare seeking for childhood illness (96.9%). Women who were not working (92.6%), those of the Akan ethnic group (93.0%), those who were cohabiting (92.5%), rural dwellers (93.3%), women with secondary/higher level of education (92.8%), those with poorer wealth quintile (94.6%), Christians (92.8%) and those with one birth (93.2%) had the greatest proportions of healthcare seeking for childhood illnesses. Similarly, women with two or more birth order children (92.7%), those whose children weighed 2.5kg and above (92.6%), women who had multiple births (92.7%) and those who had female children (93.1%) had the highest proportions of healthcare seeking for childhood illnesses.

Association between women’s autonomy in healthcare decision-making and healthcare seeking for childhood illnesses

Table 2 presents results on the association between women’s autonomy in healthcare decision-making, maternal and child characteristics and healthcare seeking for childhood illnesses. With women’s autonomy in healthcare decision-making, women who decided alone on personal healthcare had 30% reduced odds of seeking healthcare for childhood illnesses compared to those who did not decide alone [AOR = 0.70, CI = 0.51–0.97]. With age, women aged 45–49 had 69% reduced odds of seeking healthcare for childhood illnesses compared to those aged 25–29 [AOR = 0.31, CI = 0.14–0.70]. Women from the Northern and Upper West regions had 72% [AOR: 0.28, CI: 0.11–0.70] and 77% [AOR: 0.23, CI: 0.09–0.58] reduced odds of seeking healthcare for childhood illnesses respectively compared to those from the Western region.

Discussion

Evidentially, since increase in female autonomy in sub-Saharan Africa is linked to improved maternal healthcare seeking behaviour and consequent reduction in ill health of children [24,30,48,49], this study focused on the association between women’s healthcare decision-making capacity and healthcare seeking for childhood illness in Ghana. We found a relatively high prevalence of healthcare seeking for childhood illnesses among women. However, only a quarter of women in this study took healthcare decisions alone. This is in line with the findings of previous studies in Ghana [27,50].

We realised that women who take healthcare decisions alone were less likely to seek healthcare for childhood illnesses, compared to those who did not decide alone. The possible reason for the finding is that in African society, men play a paramount role in determining the health needs of a woman [51]. This is because men are considered as decision makers and those who control most of the resources in marriages and hence they decide when and where women should seek health care [52]. In this regard, women are usually not given the opportunity to visit a health facility or healthcare provider alone or to make the decision to spend money on health care [53-55]. This certainly can have serious repercussions on health in particular and self-respect in general of the women and their children. Apart from men, sometimes other family members, caretakers and friends may also be involved in the healthcare decisions of women and this limits the ability of the woman to take decisions alone [54].

Women aged 35–49 and 45–49 were less likely seek healthcare for childhood illnesses compared to those aged 25–29. In support of our finding, Ajibade et al. [32,56] found significant relationship between maternal age and health seeking behaviour, where women of similar age group as ours were less likely to seek healthcare for their sick children. It is not clear why older women might have poor healthcare seeking behaviour for sick children, but one plausible reason could be that women of this age group may have inadequate support, energy and other capacity/empowerments (such as financial, educational, relationship power dynamics) to seek healthcare. Others may be influenced by socio-cultural factors that could affect their health seeking behaviour [23,24]. Another possible reason could be that older women may have gained experience on childcare from previous children, and hence reduced their need for healthcare seeking.

Women from the Northern and Upper West regions were less likely to seek healthcare for childhood illnesses compared to those from the Western region. Consistent with our finding, Paul and Chouhan [57] in a similar study found regional differences in maternal antenatal care (ANC) visits and child health services. The reason for this finding is not far-fetched, but we infer that individual behavioural factors could be driving these regional geographical disparities in health seeking behaviour among women, especially under a standard health insurance system and ANC health policy targeted at improving women, and child health utilisation of healthcare services [58,59]. Northern Ghana (i.e. Northern, Upper West and Upper East regions) has poorer economic and health outcomes compared to other regions in the country due to geographical, historic, and socio-cultural factors that have often excluded the north from much of Ghana’s economic growth [60]. We suggest further studies that could involve a mixed method approach to better understand these regional differentials in healthcare seeking behaviour.

Strengths and limitations

The major limitation of the study is the cross-sectional nature of the survey, which makes it impossible to draw causal interpretation between the variables; at best only associations can be drawn. Additionally, these health conditions were not based on any medical diagnosis. Data was obtained from respondents’ self-report and this has the tendency to be under-reported or over-reported. Furthermore, since this is a secondary data analyses, we could not include health system/service related factors that may be related to child healthcare seeking behaviour [3]. Despite these limitations, the study has been able to unearth association between women’s autonomy in healthcare decision-making and healthcare seeking behaviour for childhood illness using nationally representative dataset. This makes the conclusions generalisable to women in Ghana. The relatively large sample size used in this study made it possible to employ rigorous statistical analysis that makes the findings and conclusions from this study valid.

Conclusion

Ghanaian women with autonomy in healthcare decision-making, those who were older and from Northern and Upper West regions were less likely to seek healthcare for childhood illness. To reduce childhood mortalities and morbidities in Ghana, we recommend educating women who take healthcare decisions alone, older women and women from deprived regions such as Northern and Upper West regions on the importance of seeking healthcare for childhood illnesses. It is also imperative to conduct a qualitative study to unravel the nuances surrounding women’s autonomy in healthcare decision making and healthcare seeking for childhood illnesses.

STROBE 2007 (v4) statement—Checklist of items that should be included in reports of cross-sectional studies.

(DOCX)

Click here for additional data file.

25 May 2020

PONE-D-20-07266

Women’s autonomy in healthcare decision-making and healthcare seeking behaviour for childhood illness in Ghana: Analysis of data from the 2014 Ghana Demographic and Health Survey

PLOS ONE

Dear Dr. Abdul-Aziz Seidu,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by 20 June. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Sharon Mary Brownie

Academic Editor

PLOS ONE

Editor Comments:

Please clarify the following comment before your manuscript is sent out for review.

......'women whose children weighed 50-100 kg, and 101-150kg were more likely to seek healthcare for childhood illnesses'

These weight ranges are not commensurate with childhood weight averages

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your Methods section, please provide additional information about the  demographic details of your participants. In paritcluar, please clarify how "child weight" was categorised, and ensure that the numbers reported are correct, as it seems unlikely that children aged 0-5 years weighted up to 200 kg.

3. Please correct your reference to "p=0.000" to "p<0.001" or as similarly appropriate, as p values cannot equal zero.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

31 May 2020

Abdul-Aziz Seidu

University of Cape Coast, Ghana

Department of Population and Health

31/05/2020

Dear Editor,

Thank you for giving us the opportunity to revise our manuscript before sending it out for peer review. We have revised it based on the issues raised.

1. Comment: ......'women whose children weighed 50-100 kg, and 101-150kg were more likely to seek healthcare for childhood illnesses'

Response: Please we have corrected this. We have corrected to “Child’s weight was recoded into below 2.5kg (1) and 2.5kg and above (2). This weight was generated based on studies that have considered a child’s weight less than 2.5kg as low birth weight and those 2.5kg and above as non-low birth weight [35-38].

2. Comment: These weight ranges are not commensurate with childhood weight averages

Response: Child’s weight was recoded into below 2.5kg (1) and 2.5kg and above (2). This weight was generated based on studies that have considered a child’s weight less than 2.5kg as low birth weight and those 2.5kg and above as non-low birth weight [35-38].

Comment: 3. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Response: Please we have formatted the manuscript to met PLOS ONE's style requirements

Comment: 4. In your Methods section, please provide additional information about the demographic details of your participants. In particular, please clarify how "child weight" was categorised, and ensure that the numbers reported are correct, as it seems unlikely that children aged 0-5 years weighted up to 200 kg.

Response: Child’s weight was recoded into below 2.5kg (1) and 2.5kg and above (2). This weight was generated based on studies that have considered a child’s weight less than 2.5kg as low birth weight and those 2.5kg and above as non-low birth weight [35-38].

Comment: 5. Please correct your reference to "p=0.000" to "p<0.001" or as similarly appropriate, as p values cannot equal zero.

Response: Please this has been corrected. Thank you.

Yours Sincerely,

On behalf of the authors

Abdul-Aziz Seidu

Submitted filename: Response to comments.docx

Click here for additional data file.

3 Sep 2020

PONE-D-20-07266R1

Women’s autonomy in healthcare decision-making and healthcare seeking behaviour for childhood illness in Ghana: Analysis of data from the 2014 Ghana Demographic and Health Survey

PLOS ONE

Dear Dr. Abdul-Aziz Seidu,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by 5th October 2020. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Sharon Mary Brownie

Academic Editor

PLOS ONE

Editor Comments

Reviewers have provided some comprehensive feedback. Please consider these carefully and respond to each recommendation.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Overall comment

The study assessed the relationship between women’s autonomy in healthcare decision-making and their healthcare seeking behaviour for childhood illnesses in Ghana. It highlights that healthcare seeking behaviour is good with >90% seeking care and that only 7% of Ghanaian women have autonomy in healthcare decision-making. Women autonomy could be a double-edged sword reflecting women empowerment as well as be a barrier to healthcare, as highlighted in this study where autonomous women were less likely to seek care. While the findings are key in ensuring that "no one is left behind" in Ghana, the study requires major revisions, which are outlined below.

Abstract

1. Revise the stated objective to make it specific and measurable and consistently use it throughout the manuscript (see lines 94-95)

2. Be specific about what statistical analyses were performed and why. Descriptive and inferential statistics is too general and does not point out how the research questions were answered.

3. The abstract results show provide some descriptive findings to provide context for a reader e.g. include information on sample characteristic, % of women seeking healthcare for childhood illness and % of women who are autonomous in decision making

Introduction

4. The introduction is focused on global/regional context. Local (Ghana) context is missing despite available studies exploring acute childhood illnesses, women autonomy, healthcare decision-making and health seeking behaviours including the Ghana Demographic Health Surveys

5. It is not clear from the introduction, what the research problem is. While authors note that there is absence of studies on the subject, the authors have argued why it is important for the subject to be studied and how the specific linkage, in particular, is important in Ghana.

6. Line 57 and 68: The authors refer to low- and middle-income regions. This reference is too broad and it is not clear whether there are regions that are LMI but they are countries within a region that are HIC, MIC or LIC. I suggest they use low- and middle-income countries, which is more specific.

7. The authors should correct reference their work e.g. Line 76, the authors reference a single Ethiopian studies in a general statement about developing countries; Lines 57-59, references 1 and 7 refer to WHO yet the cited information does not correctly represent the information in the references [The authors cite low- and middle-regions yet the reference talks of Sub-Saharan Africa and Central and South Asia].

Materials and Methods

8. Line 107: How many households were selected?

9. The authors should provide a clear breakdown of how the study sample was arrived at e.g. Out of the 11835 HH interviewed in the survey, how many women were interviewed and how many of those women had children ≤5 years and how many of them were married. It would also important to know the total number of children included

10. Outcome variables: The authors should clarify how they dealt with the women with children who did not have diarrhoea, fever, or cough in the last 2 weeks.

11. What was the conceptual definition of “women’s autonomy in healthcare decision-making”? How did the authors deal with joint decision making and decisions by other people e.g. caretakers etc.? The limitations of the key independent variable should be discussed e.g. Seymour and Peterman (2017), Understanding the Measurement of Women’s Autonomy Illustrations from Bangladesh and Ghana (http://ebrary.ifpri.org/utils/getfile/collection/p15738coll2/id/131367/filename/131578.pdf) discusses these limitations.

12. All the study variables should be clearly operationalised and source reference included. I suggest the author reference specific studies where each study variable was obtained from instead of the general statement “from previous studies”

13. I suggest line 139-140 be deleted and lines 137-139 be moved much earlier in the section.

14. Child weight: Are the authors referring to birth weight or the child weight during the survey? If it is the former, how is it related with childhood illness / healthcare seeking today? What was the rationale for categorisation of the continuous variable – child weight? Why did the authors categorise it into two instead of three (LBW, Normal, Overweight)?

15. It is not clear how the authors recoded the various independent variables e.g. education is recoded up to secondary education while occupation is recoded as working/not working. Were there women with tertiary (college/university) education? What does no education mean – no formal education or? For occupation, what does ‘working’ refers to? Are homemakers included? How were women on self-employment classified?

16. What informed the categorisation of age? Why was age not used as continuous variable or categorised into 10 years age-group? What informed the choice of 25-29 years as the reference category?

17. The use of the term influence throughout the manuscript may be construed to infer causation, yet these is an association study. I suggest the authors revise the manuscript (use association/relationship instead) to reflect the same.

18. The description of Model II and III is incorrect. These are multivariable logistic regressions and not binary logistic regression as stated. Also, the purpose of the models is to adjust for possible confounders and assess how they affect the relationship between healthcare-seeking and women’s autonomy in decision-making.

19. The use of variance inflation factor (VIF) in this study is not clear. All the study variables are categorical. My understanding is that VIF cannot be use with categorical variable because it is suitable with variables having 1 degree of freedom (which is not the case of categorical variable). Instead, generalized VIF can be used for categorical variables in R [not sure in Stata]

20. The method section should be referenced appropriately – Information on the data, study variables e.g. questions and ethical approvals should be referenced to the DHS

21. Could the inclusion of all the variables have affected the findings? Was the final model the best fit or the most robust model? I suggest the authors consider reviewing the inclusion of variables in the final model (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2633005/ and https://mybiostats.files.wordpress.com/2015/03/model_building_strategies_and_methods_for_logistic_regression.pdf)

22. I suggest that the authors use the STROBE reporting guidelines

Results

23. The authors should consider presenting percentages up to one decimal place. The additional decimal place does not improve precision of the findings. Also consider making the findings clear and concise.

24. Suggest that the authors organise their results using the following sub-titles: (a) Sample characteristics (b) Healthcare-seeking for childhood illness (c) Women’s autonomy in healthcare decision-making (d) Association between women’s autonomy in healthcare decision-making and healthcare seeking for childhood illnesses. The tables could be: Table 1. Sample characteristics; Table 2. Women’s autonomy in healthcare decision-making and Healthcare-seeking for childhood illness according to sample characteristics; Table 3: Binary and multivariable logistic regression analysis (Model 1: Bivariate Model 2: Adjust for maternal Model 3: Adjust for maternal and child characteristics

25. The authors should revise the interpretation of the odds ratio e.g. Women from the Northern and Upper West regions had 72% [AOR: 0.28, CI: 0.11-0.70] and 73% [AOR: 0.23, CI: 0.09-0.58] reduced odds of seeking healthcare for childhood illnesses respectively compared to those from the Western region.

Discussion

26. Line 220-222: The study findings do not support the statement “…issues of maternal utilization of these services limited by female autonomy in healthcare decision-making”. The study findings show that women autonomy is in factor a barrier to healthcare seeking behavior and that the level of healthcare seeking for childhood illness is very high in Ghana.

27. From the study findings, there is good health seeking practices for childhood illness regardless of the maternal and child characteristics. This should be discussed. The characteristics of women who were autonomous (made decisions alone) would help understand health seeking behaviours.

28. What is the implication on the study of the high proportion (72.3%) of women who did not decide on their healthcare alone? This should be discussed.

29. The study findings showed that making decision alone reduced the odds of healthcare seeking for childhood illness by 30%. The argument line 228-238 is focussed on limitation of women to make their own decision and how it affects their health-seeking behaviours. However, it does not address why the “women who made decision alone were less likely to seek healthcare for the childhood illness”. Importantly, the study findings indicate that healthcare seeking is not a problem in Ghana with >90% of women seeking care for childhood illness hence the argument that “low status of women can hinder them from recognizing and voicing their concerns about health needs even when it comes to seeking healthcare for their children” may not be valid. Also, the study does not explore the characteristics of women who make decisions alone, hence it is difficult to support these lines of discussion.

30. Line 230: Reference is made to men only yet decision making involves other people such as the extended family or caretakers

31. Line 239-247: The discussion is focussed on the 45-49 year but does not include the 35-39 years who also were less likely to seek care. The discussions should be made with reference to the women 25-29 years (are they more likely to seek care?). Is it possible that older women could also have gained experience on childcare (from previous children), hence reduced need for healthcare seeking?

Conclusion

32. I suggest the authors revise their conclusion to be in line with the study aim. E.g. Ghanaian women with autonomy in healthcare decision-making, those who were older and from Northern and Upper West regions were less likely to seek healthcare for childhood illness.

33. Revise the abstract conclusion to reflect the suggested revision in the conclusion.

34. The authors should be specific on the significance of their study, for instance, the proposed recommendation is too broad and not specific.

Major: English Editing

35. The work requires English editing to improve the clarity of some of the sentences and correct typographical and grammatical errors.

36. Revise the statement lines 55-57 for clarity. Include the specific year for the 12.7 million

37. Revise line 57-59 for clarity and to be specific. It is currently too general and unclear.

38. Lines 57-90 can be summarised into a single paragraph to reduce repetition and provide a clear and concise argument.

39. The authors should consider using low resource settings/countries instead of “developing countries”

40. Revise the statement lines 55-57 for clarity. Include the specific year for the 12.7 million

41. Line 101: Revise to reflect the correct state: The survey has so far been carried out every five-years not supposed

42. Revise “significant influence” to “association”

43. Delete line 177-178: “The data can…” This information is already stated under the data availability section.

44. Revise Line 182-183 to remove “In terms of….participants,” which is redundant.

45. Revise line 190-191 for clarity.

46. Line 215: Table 2: Revise “influence” to “Association”

47. Line 264 Include self-report

48. Revise Line 270-271 for clarity

49. Delete “In conclusion...” in line 44 and 275

Reviewer #2: The authors have addressed all comments.

However, there remains residual minor revisions which are as below:

1. The sentence on page 5 line 101, that reads “The GDHS is a nationwide survey that is supposed to be carried out every five years since it began.” The authors can state that the GDHS is a nationwide representative survey conducted every five years and if they want to add when it began they should state the year of the first survey.

2. The sentence on page 14 line 261, that reads “ This study’s major limitation is its cross-sectional which preludes causality” should be revised. It appears incomplete and hanging.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Samwel Gatimu

Reviewer #2: No

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

12 Oct 2020

Abdul-Aziz Seidu

University of Cape Coast, Ghana

Department of Population and Health

11/10/2020

Dear Editor,

AUTHOR’S RESPONSE TO REVIEWS:

Dear Editor,

This letter is in reference to your email with reviewers’ comments. We are very pleased that the manuscript is potentially acceptable for publication in PLOS ONE once we have carried out some revisions. We would like to thank the reviewers for their insightful and helpful comments and for giving us the chance to revise our manuscript. We believe the revised manuscript has been significantly improved and the reviewers’ comments have been addressed adequately. We think in its current form it will make a valuable contribution to the literature on this increasingly important topic. Please find for your kind consideration the following: 1) a section-by-section response to the comments and suggestions of the reviewers (below) and 2) the revised manuscript provided as a marked-up copy and a clean copy. All changes have been marked in Yellow. We hope that these changes meet with your favourable consideration. Please do not hesitate to get in touch if you require any further information.

Reviewer #1: Overall comment

The study assessed the relationship between women’s autonomy in healthcare decision-making and their healthcare seeking behaviour for childhood illnesses in Ghana. It highlights that healthcare seeking behaviour is good with >90% seeking care and that only 7% of Ghanaian women have autonomy in healthcare decision-making. Women autonomy could be a double-edged sword reflecting women empowerment as well as be a barrier to healthcare, as highlighted in this study where autonomous women were less likely to seek care. While the findings are key in ensuring that "no one is left behind" in Ghana, the study requires major revisions, which are outlined below.

Abstract

1. Revise the stated objective to make it specific and measurable and consistently use it throughout the manuscript (see lines 94-95)

Response: In this study, we sought to examine the association between women’s’ healthcare healthcare decision-making capacity and their healthcare seeking behaviour for childhood illnesses in Ghana. We have made this clear throughout the paper. (see line 28-30; 111-113)

2. Be specific about what statistical analyses were performed and why. Descriptive and inferential statistics is too general and does not point out how the research questions were answered.

Response: We have made the statistical analyses more clear by stating that “Chi-square test of independence and binary logistic regression were carried out to generate the results. Statistical significance was pegged at 95% confidence intervals (CIs).” (see line 34-36)

3. The abstract results show provide some descriptive findings to provide context for a reader e.g. include information on sample characteristic, % of women seeking healthcare for childhood illness and % of women who are autonomous in decision making.

Response: We have added this “Out of the 2,450 women, approximately 25.7% could take healthcare decisions alone and 89.7% sought healthcare for their childhood illnesses (see line 39-40).

Introduction

4. The introduction is focused on global/regional context. Local (Ghana) context is missing despite available studies exploring acute childhood illnesses, women autonomy, healthcare decision-making and health seeking behaviours including the Ghana Demographic Health Surveys

Response: We have added studies in Ghana to the background (see line 66-70, 88-93).

5. It is not clear from the introduction, what the research problem is. While authors note that there is absence of studies on the subject, the authors have not argued why it is important for the subject to be studied and how the specific linkage, in particular, is important in Ghana.

Response: We have made clear the statement of the problem by justifying why the study is important in Ghana. See line 105-111.

6. Line 57 and 68: The authors refer to low- and middle-income regions. This reference is too broad and it is not clear whether there are regions that are LMI but they are countries within a region that are HIC, MIC or LIC. I suggest they use low- and middle-income countries, which is more specific.

Response: This has been revised. We have replaced low-and middle-income regions with low-and middle-income countries (see line 83 and 94).

7. The authors should correct reference their work e.g. Line 76, the authors reference a single Ethiopian studies in a general statement about developing countries; Lines 57-59, references 1 and 7 refer to WHO yet the cited information does not correctly represent the information in the references [The authors cite low- and middle-regions yet the reference talks of Sub-Saharan Africa and Central and South Asia].

Response: We have revised this section of the paper (see line 64-66)

Materials and Methods

8. Line 107: How many households were selected?

Response: We have given the number of households as 12831 in the methods (see line 126)

9. The authors should provide a clear breakdown of how the study sample was arrived at e.g. Out of the 11835 HH interviewed in the survey, how many women were interviewed and how many of those women had children ≤5 years and how many of them were married. It would also important to know the total number of children included.

Response: A breakdown of the sample has been provided in the methods (see line 124-130)

10. Outcome variables: The authors should clarify how they dealt with the women with children who did not have diarrhoea, fever, or cough in the last 2 weeks.

Response: We have made this clear by stating “In this study, only childbearing women whose children had either diarrhoea/cough/fever in the last 2 weeks were considered. Hence, childbearing women whose children did not have diarrhoea/cough/fever in the last 2 weeks were excluded from the study (see line 127-130).

11. What was the conceptual definition of “women’s autonomy in healthcare decision-making”? How did the authors deal with joint decision making and decisions by other people e.g. caretakers etc.? The limitations of the key independent variable should be discussed e.g. Seymour and Peterman (2017), Understanding the Measurement of Women’s Autonomy Illustrations from Bangladesh and Ghana (http://ebrary.ifpri.org/utils/getfile/collection/p15738coll2/id/131367/filename/131578.pdf) discusses these limitations.

Response: We have clearly indicated how women’s autonomy in healthcare decision-making was conceptualised and acknowledged the limitation of the variable (see line 151-158).

12. All the study variables should be clearly operationalised and source reference included. I suggest the author reference specific studies where each study variable was obtained from instead of the general statement “from previous studies”

Response: We have provided specific references to all the variables associated with healthcare seeking for childhood illnesses (see line 158-163).

13. I suggest line 139-140 be deleted and lines 137-139 be moved much earlier in the section.

Response: We have deleted lines 139-140 and moved line 137-139 much earlier in the section. (see line 158-160).

14. Child weight: Are the authors referring to birth weight or the child weight during the survey? If it is the former, how is it related with childhood illness / healthcare seeking today? What was the rationale for categorisation of the continuous variable – child weight? Why did the authors categorise it into two instead of three (LBW, Normal, Overweight)?

Response: Child weight is the weight of the child at birth. It is related because, the weight of the child at birth is associated with childhood illnesses (fever, diarhea and cough) and will therefore play a role in determining the mothers’ healthcare seeking as the child grows. The reason for the two categories (low birth weight versus non-low birth weight) was to assess how low birth weight or otherwise is linked to healthcare seeking for childhood illnesses (see line 169-170).

15. It is not clear how the authors recoded the various independent variables e.g. education is recoded up to secondary education while occupation is recoded as working/not working. Were there women with tertiary (college/university) education? What does no education mean – no formal education or? For occupation, what does ‘working’ refers to? Are homemakers included? How were women on self-employment classified?

Response: With education, the actual coding was no formal education, primary and secondary/higher (tertiary). Higher was mistakenly left out in the methods but was included in the analysis. We have corrected this in the methods. With occupation, working refers to respondents engaged in occupations such as managerial, clerical, sales, agriculture, services and manual. These were obtained from the dataset (see line 164-166)

16. What informed the categorisation of age? Why was age not used as continuous variable or categorised into 10 years age-group? What informed the choice of 25-29 years as the reference category?

Response: We did not recode age. In the standard demographic and health survey data that is how age is categorised and the choice of references were based on categories that had high prevalence of healthcare seeking behaviour

17. The use of the term influence throughout the manuscript may be construed to infer causation, yet these is an association study. I suggest the authors revise the manuscript (use association/relationship instead) to reflect the same.

Response: We have revised taken out influence from appropriate aspects of the paper and replaced it with association/relationship.

18. The description of Model II and III is incorrect. These are multivariable logistic regressions and not binary logistic regression as stated. Also, the purpose of the models is to adjust for possible confounders and assess how they affect the relationship between healthcare-seeking and women’s autonomy in decision-making.

Response: We have revised this section under statistical analysis (see line 185-191)

19. The use of variance inflation factor (VIF) in this study is not clear. All the study variables are categorical. My understanding is that VIF cannot be use with categorical variable because it is suitable with variables having 1 degree of freedom (which is not the case of categorical variable). Instead, generalized VIF can be used for categorical variables in R [not sure in Stata].

Response: Thanks for your comment. In STATA, VIF can be generated for categorical variables

20. The method section should be referenced appropriately – Information on the data, study variables e.g. questions and ethical approvals should be referenced to the DHS

Response: Thank you for your comment. We have referenced information under the methods section to DHS.

21. Could the inclusion of all the variables have affected the findings? Was the final model the best fit or the most robust model? I suggest the authors consider reviewing the inclusion of variables in the final model (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2633005/ and https://mybiostats.files.wordpress.com/2015/03/model_building_strategies_and_methods_for_logistic_regression.pdf)

Response: The variable selection was informed by variables that were significant at the bivariate level, and the model fitness was informed by the values of the pseudo R2, which was highest in the final model.

22. I suggest that the authors use the STROBE reporting guidelines

Response: We have used this and attached it as a supplementary file (see line 197-199).

Results

23. The authors should consider presenting percentages up to one decimal place. The additional decimal place does not improve precision of the findings. Also consider making the findings clear and concise.

Response: We have revised the results section and presented the percentages up to one decimal place (see line 208-216; 219-231)

24. Suggest that the authors organise their results using the following sub-titles: (a) Sample characteristics (b) Healthcare-seeking for childhood illness (c) Women’s autonomy in healthcare decision-making (d) Association between women’s autonomy in healthcare decision-making and healthcare seeking for childhood illnesses. The tables could be: Table 1. Sample characteristics; Table 2. Women’s autonomy in healthcare decision-making and Healthcare-seeking for childhood illness according to sample characteristics; Table 3: Binary and multivariable logistic regression analysis (Model 1: Bivariate Model 2: Adjust for maternal Model 3: Adjust for maternal and child characteristics.

Response: We have revised these sections per your suggestion but with slight modification. For instance, instead of having two separate tables for Sample characteristics and women’s autonomy in healthcare decision-making and Healthcare-seeking for childhood illness according to sample characteristics (Table 1 and 2), we think presenting the information in one table like we already did is better.

25. The authors should revise the interpretation of the odds ratio e.g. Women from the Northern and Upper West regions had 72% [AOR: 0.28, CI: 0.11-0.70] and 73% [AOR: 0.23, CI: 0.09-0.58] reduced odds of seeking healthcare for childhood illnesses respectively compared to those from the Western region.

Response: We have revised the interpretation of the odds ratio (see line 237-245).

Discussion

26. Line 220-222: The study findings do not support the statement “…issues of maternal utilization of these services limited by female autonomy in healthcare decision-making”. The study findings show that women autonomy is in factor a barrier to healthcare seeking behavior and that the level of healthcare seeking for childhood illness is very high in Ghana.

Response: We have taken this out of the discussion.

27. From the study findings, there is good health seeking practices for childhood illness regardless of the maternal and child characteristics. This should be discussed. The characteristics of women who were autonomous (made decisions alone) would help understand health seeking behaviours.

Response: We have discussed the prevalence of healthcare seeking behaviour (see line 257-266).

28. What is the implication on the study of the high proportion (72.3%) of women who did not decide on their healthcare alone? This should be discussed.

Response: We have discussed and given the implications of the high proportion of women who did not decide on their healthcare alone (see line 257-266).

29. The study findings showed that making decision alone reduced the odds of healthcare seeking for childhood illness by 30%. The argument line 228-238 is focussed on limitation of women to make their own decision and how it affects their health-seeking behaviours. However, it does not address why the “women who made decision alone were less likely to seek healthcare for the childhood illness”. Importantly, the study findings indicate that healthcare seeking is not a problem in Ghana with >90% of women seeking care for childhood illness hence the argument that “low status of women can hinder them from recognizing and voicing their concerns about health needs even when it comes to seeking healthcare for their children” may not be valid. Also, the study does not explore the characteristics of women who make decisions alone, hence it is difficult to support these lines of discussion.

Response: We have taken out the invalid sentences, making the paragraph focus only on reasons why women who made decision alone were less likely to seek healthcare for the childhood illness (see line 267-277).

30. Line 230: Reference is made to men only yet decision making involves other people such as the extended family or caretakers.

Response:: We have added a sentence to indicate that sometimes other family members, caretakers and friends are involved in decision making and this limits the ability of the woman to take decisions alone (see line 275-277).

31. Line 239-247: The discussion is focussed on the 45-49 year but does not include the 35-39 years who also were less likely to seek care. The discussions should be made with reference to the women 25-29 years (are they more likely to seek care?). Is it possible that older women could also have gained experience on childcare (from previous children), hence reduced need for healthcare seeking?

Response: We have revised this aspect of the discussion (see line 278-288).

Conclusion

32. I suggest the authors revise their conclusion to be in line with the study aim. E.g. Ghanaian women with autonomy in healthcare decision-making, those who were older and from Northern and Upper West regions were less likely to seek healthcare for childhood illness.

Response: We have revised the conclusion to read “Ghanaian women with autonomy in healthcare decision-making, those who were older and from Northern and Upper West regions were less likely to seek healthcare for childhood illness” (see line 315-316)

33. Revise the abstract conclusion to reflect the suggested revision in the conclusion.

Response: We have revised this (see line 50-55)

34. The authors should be specific on the significance of their study, for instance, the proposed recommendation is too broad and not specific.

Response: We have revised our recommendation to read “To reduce childhood mortalities and morbidities in Ghana, we recommend educating women such as older women, women from deprived regions such as Northern and Upper West regions on the need to involve others in their healthcare decision making so as to enhance their healthcare seeking for childhood illnesses.” (see line 315-320).

Major: English Editing

35. The work requires English editing to improve the clarity of some of the sentences and correct typographical and grammatical errors.

Response: We have proof read the paper to correct the grammatical errors.

36. Revise the statement lines 55-57 for clarity. Include the specific year for the 12.7 million

Response: We have clarified this. See line 64

37. Revise line 57-59 for clarity and to be specific. It is currently too general and unclear.

Response: We have revised this sentence. See line 64-65

38. Lines 57-90 can be summarised into a single paragraph to reduce repetition and provide a clear and concise argument.

Response: Thank you for your suggestion. The lines 57-90 have been developed into different paragraphs that communicate different ideas. Instead of summarising into a single paragraph, we have revise the sentences to make them clear and concise.

39. The authors should consider using low resource settings/countries instead of “developing countries”

Response: We have revised this to low and middle-income countries. See line 64 and 83.

41. Line 101: Revise to reflect the correct state: The survey has so far been carried out every five-years not supposed.

Response: We have revised this. See line 118-119

42. Revise “significant influence” to “association”

Response: We. Have revised influence to association in the entire manuscript

43. Delete line 177-178: “The data can…” This information is already stated under the data availability section.

Response: We have deleted the information.

44. Revise Line 182-183 to remove “In terms of….participants,” which is redundant.

Response. We revised the sentence.

45. Revise line 190-191 for clarity.

Response. We revised the sentence. See line 219-220.

46. Line 215: Table 2: Revise “influence” to “Association”

Response: We. Have revised influence to association in the entire manuscript

47. Line 264 Include self-report

Response: We have added self-report. See line 304-306.

48. Revise Line 270-271 for clarity

Response: We have revised the sentence. See line 331-332

49. Delete “In conclusion...” in line 44 and 275

Response: We have deleted the phrase.

Reviewer #2: The authors have addressed all comments.

However, there remains residual minor revisions which are as below:

1. The sentence on page 5 line 101, that reads “The GDHS is a nationwide survey that is supposed to be carried out every five years since it began.” The authors can state that the GDHS is a nationwide representative survey conducted every five years and if they want to add when it began they should state the year of the first survey.

Response. This has been revised. See line 118-119

2. The sentence on page 14 line 261, that reads “ This study’s major limitation is its cross-sectional which preludes causality” should be revised. It appears incomplete and hanging.

Response: We have revised the sentence. See line 302-304

Yours Sincerely,

On behalf of the authors

Abdul-Aziz Seidu

Submitted filename: Response to comments.docx

Click here for additional data file.

16 Oct 2020

Women’s autonomy in healthcare decision-making and healthcare seeking behaviour for childhood illness in Ghana: Analysis of data from the 2014 Ghana Demographic and Health Survey

PONE-D-20-07266R2

Dear Dr. Abdul-Aziz Seidu,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Sharon Mary Brownie

Academic Editor

PLOS ONE

Editor Comments

Reviewers comments have been addressed.

26 Oct 2020

PONE-D-20-07266R2

Women’s autonomy in healthcare decision-making and healthcare seeking behaviour for childhood illness in Ghana: Analysis of data from the 2014 Ghana Demographic and Health Survey

Dear Dr. Seidu:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Sharon Mary Brownie

Academic Editor

PLOS ONE