Expectant fathers’ knowledge of maternal morbidity: a Sri Lankan experience

Background: Male partners play an important and vital role in the decision-making process regarding pregnant women’s health. The purpose of the present study was to assess the knowledge and awareness of expectant fathers about Gestational Diabetes Mellitus (GDM), Pregnancy Induced Hypertension (PIH), and anaemia during pregnancy. Methods: A cross sectional descriptive study was carried out among expectant fathers whose partners were attending antenatal clinics at the Anuradhapura Teaching Hospital, Sri Lanka. All consenting participants were interviewed by investigators using an interviewer administered questionnaire to collect data on knowledge of risk factors, symptoms, complications and their control. Statistical analysis was performed using the Kruskal Wallis test. Results: Of the 246 expectant fathers studied, 192 (78%) were aware of GDM, 183 (74.4%) and 154 (62.6%) were aware of PIH and anaemia during pregnancy, respectively. The total number of answers provided by expectant fathers ranged from 0 to 33 (of 41 questions). There were 44 fathers who could not answer even a single question. For GDM, anaemia, and PIH, the percentages of expectant fathers who failed to provide at least a single correct answer were 24.8%, 40.2%, and 31.3%, respectively. The median number of total correct answers provided increased steadily along with the average income (chi-square 31.24, p<0.001) and educational level (chi-square 33.57, p<0.001). Expectant fathers in the 25-34 age group had significantly higher scores, compared to younger and older fathers (chi-square 15.11, p=0.001). Fathers experiencing the second pregnancy of their spouses also had higher scores. Conclusions: Expectant father’s knowledge of the selected morbidities was limited. To improve maternal health, any health promotional programmes should include expectant fathers.

The important role of men as the primary stakeholders of women's reproductive and sexual health has been widely accepted. In 1994, 20,000 United Nations delegates with member status pressed for men's involvement in reproductive health. The programme of action of the International Conference on Population and Development (ICPD) stressed men's shared responsibility for women's health 1 . Although the primary focus was on reducing violence against women and children, the resolution clearly declared that men should take responsibility for the empowerment of women.
Studies on family planning 2,3 , HIV/AIDS 4 , abortions 4 and breastfeeding 5 all provide clear evidence for the important and vital role of male partners in the decision-making process regarding women's health. Interventional studies on reproductive health programmes targeting either males or both sexes, compared to programmes targeting only females, have shown a significant effect on the reproductive health outcome of women. With this growing body of evidence, the concept of men as "gatekeepers" of women's health has been developed into a positive approach that includes men as partners in improving women's health.
Nevertheless, these involvements are often driven by the gender power relationships given to male partners in various cultures 6 . While reproductive health education and health promotion programmes that target adolescent boys and young men are acceptable in most parts of the world, shared responsibility for pregnancyrelated matters heavily depends on the nature of gender roles within the society. In addition, fear of losing respect from peers, lack of communication skills, lack of knowledge, and perceptions of masculinity are also shown to have a major effect on this involvement. Despite having a lack of knowledge and involvement, men often dominate the decision-making processes related to pregnancy, especially in male-dominated South Asian cultures.
Studies in maternal health have shown that early interventions during the antenatal period have a major protective effect for medical conditions complicating pregnancy and direct pregnancy-related acute complications such as haemorrhage, rupture of the uterus, and obstructed labour. These interventions include early identification, early treatment and behaviour modifications.
However, screening tests and procedures for these conditions require resources, and early detection is not currently at an optimal level in most developing countries. We hypothesized that this could be partly due to a lack of awareness about maternal morbidity among pregnant women, and specifically among their partners who play a major role in decision-making. The knowledge of expectant fathers is vital to ensure that healthcare is sought early to prevent complications due to these conditions. The purpose of the present study was to assess the knowledge and awareness of expectant fathers about Gestational Diabetes Mellitus (GDM), Pregnancy Induced Hypertension (PIH), and anaemia during pregnancy.

Methods
This study conformed to the Helsinki Declaration and to local legislation. All participants gave informed consent to participate in this study. Ethical clearance was obtained from the Research and Ethics Committee, Faulty of Medicine and Allied Sciences, Rajarata University of Sri Lanka.
The study was conducted in Anuradhapura, Sri Lanka, during November to December 2010. Anuradhapura district is situated in the North Central Province of Sri Lanka. The resident population is around 830,000 7 . Over the five years preceding this study the annual number of births reported from the district was around 16,000. The crude birth rate is 20.7/1000 population. Infant and neonatal mortality rates of Anuradhapura district (17.2/1000 live births (LB) and 11/1000 LB) are significantly higher than those reported in Sri Lankan national data, which are 11/1000 LB and 8.7/1000 LB, respectively 8 .
The study population for the present study included expectant fathers whose partners were attending antenatal clinics at the Anuradhapura Teaching Hospital. The study sample included the expectant fathers that were accompanying pregnant women to the antenatal clinics. The participants included fathers who were waiting during the clinic hours as well as the fathers who came to the clinic just to drop off the pregnant women at the clinic (this was a common practice and usually they returned around four hours later). The authors visited the antenatal clinics at the starting time and introduced themselves to expectant fathers, explained the objectives of the study, and handed out the self-administered questionnaires.
The study sample for the present study included all expectant fathers visiting selected clinics during the study period. Since the study followed a non-probability sampling procedure, the optimal sample size was not calculated. However, it was hypothesized that if the 80% of the expectant fathers are aware of the selected morbidities and a simple random sampling procedure was followed, a minimum of 246 expectant fathers were required with 95% confidence limits and 5% absolute precision.
Variables for the study included basic socio-demographic variables and knowledge about GDM, PIH and anaemia. For each condition selected, expectant fathers were first asked about their awareness of the existence of certain disease conditions, for example, 'Are you aware that pregnant women could develop a condition called gestational diabetes mellitus?'. Then a selected set of questions on risk factors, clinical features, complications and preventative measures (primary and secondary) was included for all selected conditions (e.g., 'Which of the following could be risk factors for developing GDM?'). Only true risk factors, features and preventative measures were provided in the questionnaire. This approach was followed as a way of including health education and to avoid the possibility of giving incorrect messages through the survey. The variables were selected after discussion with medical, obstetric, and community medicine experts. The questionnaire was developed in English and translated into Sinhalese.
All demographic variables were analyzed as categorical variables. Answers to individual questions were presented as percentages. The total number of correct answers provided by participants was also reported as a percentage. The knowledge distribution was expected to have a skewed distribution because of the distribution of educational status in this population is also skewed, and non-parametric tests were carried out for significance testing.

Results
The total number of expectant fathers studied was 246. The mean age of the participants was 30.2 years (standard deviation (SD) of 6.2 years) and the mean age of the pregnant mothers was 26.6 years (SD of 5.6 years). Table 1 shows the demographic characteristics of the study sample.
Of the risk factors listed in the questionnaire, a family history of GDM/diabetes mellitus (DM) was recognized by 48.8% of the study sample (Table 2). Of the respondents, 135 (54.9%) knew that increased frequency of urination was a common clinical presentation for GDM/DM. The probability of developing noninsulindependent diabetes mellitus (NIDDM) in later life as a complication of GDM was known to 98 (39.8%) respondents. Screening for GDM if risk factors were present was indicated as an appropriate intervention to control GDM by 30.5% (n=75) of participants.

GDM Complications
Diabetes in later in life 98 98 Delivery of an abnormally large baby 17 6.9 Diabetes in child 51 20.7 Neonatal problems 6 2.4 Obstructed labor 11 4.5

GDM Control and prevention
Perform a blood glucose test in the booking visit/ early in pregnancy 71 28.9 Perform specific blood tests if risk factors/clinical symptoms present 75 30.5 the respondents. However 92 fathers (37.4%) reported headache as a common clinical presentation of anaemia, and 65 (26.4%) were aware of the increased risk of postpartum haemorrhage. Half of the study sample responded that adding iron-rich food to the diet would help to prevent anaemia. Dizziness was identified as a probable symptom of PIH by 58.1% (n=143) participants. The most well known complication of PIH was stillbirth (13.4%). Nearly half of the participants agreed that PIH needed pharmaceutical intervention for proper control.

Anaemia Risk factors
The total number of answers provided by the expectant fathers ranged from zero to 33 (of 41 questions). The distribution of the total scores showed a highly-skewed distribution with majority aggregating towards low scores. There were 44 fathers who could not answer even a single question. For GDM, anaemia, and PIH, the percentage of expectant fathers who failed to provide at least a single correct answer was 24.8%, 40.2%, and 31.3%, respectively.
The distribution of score totals by socio-demographic characteristics is illustrated in Figure 1- Figure 5. Median scores increased steadily along with the average income and educational level. Expectant fathers in the 25-34 age group had significantly higher scores, compared to younger and older fathers. Fathers experiencing the second pregnancy of their spouses also had higher scores. Ethnic group was not associated with the scores obtained in this study sample.  ; whiskers extend 1.5 IQR from the box; circles shows outliers lying between 1.5 to 3 IQR from box and the asterisk shows extremes more than 3 IQR from box. Numbers denotes the identification number for the study unit).
Increasing maternal age was the risk factor most commonly recognized among expectant fathers as a risk factor for developing anaemia during pregnancy, which was reported by 26.8% (n=66) of the fathers. All other risk factors were known to less than 15% of ; whiskers extend 1.5 IQR from the box; circles shows outliers lying between 1.5 to 3 IQR from box and the asterisk shows extremes more than 3 IQR from box. Numbers denotes the identification number for the study unit). ; whiskers extend 1.5 IQR from the box; circles shows outliers lying between 1.5 to 3 IQR from box and the asterisk shows extremes more than 3 IQR from box. Numbers denotes the identification number for the study unit).  ; whiskers extend 1.5 IQR from the box; circles shows outliers lying between 1.5 to 3 IQR from box and the asterisk shows extremes more than 3 IQR from box. Numbers denotes the identification number for the study unit). circles shows outliers lying between 1.5 to 3 IQR from box and the asterisk shows extremes more than 3 IQR from box. Numbers denotes the identification number for the study unit).

Discussion
Pregnancy and childbirth have long been considered as a women's domain 9 by communities in which the involvement of male partners is minimal. Some studies even suggest that the number of expectant fathers seeking healthcare for their spouse's problems is lower than that of non-expectant fathers 10 . The reason for this could be due to the fact that women are more likely to persuade men to access healthcare rather than men being able to persuade women to seek help 11 . However, the intentions of both partners have shown to be a better predictor of health behavior 12 ; thus the father's knowledge is vital in health-seeking behavior during pregnancy. The present study provides evidence to show that the expectant fathers have a limited knowledge of the main maternal morbidities. As this study showed that there was very poor knowledge, the results could not be interpreted, but there is definitely room for improvement.
Anemia was assumed to be a well-known condition among Sri Lankan people due to several factors. It is a major determinant in maternal health, and the public health programme in Sri Lanka has conducted endless campaigns to prevent this condition over the last few decades. In a study done in 2009-10 period, the prevalence of anemia among pregnant women in Anuradhapura was reported as 14% 13 . However, the present study shows that in Anuradhapura, 37.4% of fathers were not even aware that anemia could be a problem during pregnancy. This raises a major concern regarding the effectiveness of the present health education programmes. Either these programmes are only directed at pregnant mothers, or the programmes have failed to convey their essential messages.
Knowledge of GDM was slightly better than for other conditions, most likely due to the increasing community prevalence of diabetes in Sri Lanka, because the knowledge of the direct complications in pregnancy associated with GDM was poor. Prevalence of GDM in Sri Lanka is increasing, and a community-based study showed that around 10.3% of pregnant mothers in Sri Lanka experience GDM 14 . The present screening programme for GDM is not functioning well, due to a lack of facilities. If the community, especially expectant fathers, can be made aware of this condition, it will change careseeking behavior and more cases will be detected and treated, thus preventing severe complications to both mother and child.
Eclampsia and PIH are listed as the second leading cause of maternal deaths in Sri Lanka. The condition accounts for a considerable proportion of hospitalizations, and the disease burden is very high. Routine blood pressure and urine albumin measurements are carried out in antenatal clinics as screening tests for PIH. In this study sample, risk factors and complications of PIH were known to a limited number of expectant fathers. However, dizziness, which is a non-specific sign of PIH, was known to more than 50% of the study sample as a clinical manifestation.
One major observation related to all three conditions was lack of knowledge about the complications of selected conditions. This lack of knowledge reduces the perception of risk due to these conditions. Risk perception is a core concept of health behavior, which is largely based on health literacy 15 . The study results suggest that the Sri Lankan maternal health programme should change their strategies in order to improve knowledge about these conditions.
Exploratory models have shown that health risk perception which is the basis for care-seeking depends on social class 16 , socioeconomic determinants, and the structure and practices of the health system 17 . Among pregnant mothers, knowledge about pregnancy risk factors 18 and GDM has been shown to be associated with educational status 19 . We also observed that level of education and income are positively associated with expectant fathers' knowledge about maternal morbidities.
Such inequalities in social factors seriously affect heath and underpin all other determinants of health. Sri Lanka has shown remarkable success in overcoming these social inequalities in reducing maternal deaths. However, the maternal morbidity pattern still reveals a high level of inequality, and poor knowledge among expectant fathers could be one major determinant of the problem.

Limitations
The study sample consisted of expectant fathers who were visiting clinics. These fathers may be more health conscious and not representative of the total population of expectant fathers. The knowledge observed may be higher than the actual prevailing knowledge. The majority of the individuals in the study sample were visiting antenatal clinics at the teaching hospital in Anuradhapura. These fathers might represent a group of fathers whose socioeconomic background was different from fathers living in rural areas. Some of the questions included in the questionnaire were testing a very high level of health literacy. However, the percentage of correct answers was only used as a subjective score in the interpretation.

Conclusions
The limited level of knowledge among expectant fathers clearly shows room for the improvement of maternal health programmes. We suggest that safe motherhood programmes should have practical and operational strategies to include expectant fathers in maternal health promotion programmes.

Consent
All participants gave informed consent to participate in this study. Ethical clearance was obtained from the Research and Ethics Committee, Faulty of Medicine and Allied Sciences, Rajarata University of Sri Lanka.
Author contributions AW, GLW, MPW and GMW are second year medical undergraduates. They conceptualized the study and carried out data collection, processing and analysis. SBA is the supervisor of the study, guided the other authors and prepared the manuscript.

Competing interests
No relevant competing interests were disclosed.

Grant information
Part of this study was funded through a grant from Maternal Health Task Force of Engender Health (Grant number: GMH-106-01).
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Shahirose Premji
Faculty of Nursing, University of Calgary, Calgary, Canada I wish to extend my appreciation for the invitation to referee this important study which examines expectant fathers' knowledge and awareness about pregnancy complications -gestational diabetes mellitus, pregnancy-induced hypertension, and anaemia -that may impact health outcomes of both the mother and baby. The relevance of the work undertaken is well articulated in the introduction and the authors used a non-experimental descriptive research design to assess expectant fathers' knowledge and awareness about selected morbidities. Fathers were recruited from antenatal clinics at the Anuradhapura Teaching Hospital. Eighteen percent (44 of 246) of expectant fathers were unable to answer a single question (awareness and knowledge). The remaining fathers (82%) demonstrated varied levels of awareness about gestational diabetes mellitus (78%), pregnancy-induced hypertension (74%), and anaemia (63%) and knowledge regarding risk factors, signs and symptoms, associated complications, and management strategies. The findings have implications for Sri Lankan maternal health programme intent to improve health literacy of the population and strategies to promote health seeking behaviors.

Introduction:
Hypothesis: The statement leading to the hypothesis emphasizes health inequities (i.e. lack of resources related to screening and managing morbidities); however this is not the focus of the study. Moreover, the hypothesis would suggest that the authors wish to explain why some expectant mothers and fathers seek care while other don't; asserting that lack of knowledge and awareness may be one factor. The way in which the hypothesis is stated would suggest that the study examines both expectant mothers' and expectant fathers' knowledge and awareness about selected morbidities. The descriptive nature of the study would suggest that you want to determine whether expectant fathers whose partners are seeking care have increased knowledge and awareness given their role in decision making about seeking care.

Design and Methods:
Setting: Provide a more detailed description of the study setting to determine how representative the sample is of the population seeking care at this hospital. For example, whether it is common practice for expectant fathers to accompany their partner (i.e. wait with then or drop them off) to the antenatal clinic, and the average number of women who attend the clinic per day, to give perspective for the short duration of recruitment (November to December 2010).

Recruitment:
How did you enroll the expectant fathers who dropped off their partners at the clinic?
Recruitment: How did you enroll the expectant fathers who dropped off their partners at the clinic? Your demographic description should indicate the numbers of expectant fathers who were waiting with their partners and the number who were dropping their partner off. Provide more detail related to the recruitment process (e.g. the number who refused to participate, language use, if unable to sign how was this managed, what were potential risks/harms and how did you manage these?).
Questionnaire: Provide more details. The English version of the questionnaire uses medical terms (e.g., gestational diabetes mellitus, Polyphagia) which participants may not be familiar with. For example, in some cultures diabetes is referred to as sugar in the blood. When translating the English to Sinhalese, was the language appropriately changed? What was the level of readability in the Sinhalese language? Was the questionnaire back-translated to English to ensure language equivalency? Was the questionnaire piloted? Given the variability in literacy (i.e., levels of education) within the community, how did participants complete the questionnaire? For instance, did the researcher or research assistant assist by using an interview format? On page 3 it is stated that "The total number of answers provided by the expectant fathers ranged from zero to 33 (of 41 questions)". The questionnaire shared in Appendix 1 only has 23 questions. Explain how each of the questions, particularly those with multiple components (e.g., risk factors) was presented and how responses were elicited.

Analysis:
You indicate "All demographic variables were analysed as categorical variables"; however report the mean age of participants and their partners. Were data normally distributed to report means? Revise this statement accordingly.

Results:
First paragraph: provide a brief description of your sample by highlighted distinct characteristics. For instance, for most women this was their first pregnancy. In the discussion comment on how your sample compares to the general population (i.e., how representative is it). Is it mostly first time mothers who access antenatal clinics?
Did all expectant fathers provide demographic information? If so, include n of the sample in the table.
You provide data on the prevalence of gestational diabetes mellitus, pregnancy induced hypertension, and anaemia as reported by fathers (page 3 second paragraph of results). There are two issues with this statement: (1) prevalence refers to the proportion of a population. Your study used a convenience sample.
(2) The estimate was based on fathers' report but many of the fathers were not able to recognize these morbidities. Please clarify.
Findings: need to be presented more clearly.
The missing data needs to be accounted for in the tables and explanations provided in the text.
Please explain how the percentages were calculated. There are instances when the total percent count is over 100% (e.g. GDM risk factors) and in some instance well below 100% (e.g. GDM complications).
Did all participants answer questions related to demographic characteristics? If so, an 1.
Did all participants answer questions related to demographic characteristics? If so, an important consideration would be describing how similar or different the demographic characteristics were between respondents and non-respondents.
Examination of the distribution of responses based on specific demographic variables: Explain what you mean by "score totals"m and how this score was determined? Given the multiple comparisons being made, you are likely to find a significant difference by chance (i.e., Type 1 error). Consider the p value adjustments required for these multiple comparisons. These findings should be interpreted with caution given the nature of the questionnaire (i.e., medical terms used), the type of design, and number of participants who were able to respond.

Discussion:
The first paragraph discusses expectant fathers seeking health care and non-expectant fathers seeking health care, which is not the focus of your study. Delete.
Needs more depth and breadth by addressing: Discuss the role of parity with reference to health seeking behaviors and knowledge or health literacy.
Be careful with the inferences you are making. For example, improving knowledge will improve health seeking behavior. You findings suggest that expectant fathers seek health care for their pregnant spouse even though they don't have the knowledge.
Refocus the discussion with emphasis on the findings and implications for practice, future research and policy.

Limitations:
Consider revising after you have addressed the above points. For instance, if medical terms were used, this may explain why so many respondents were not able to complete even one question. This is an important body of work as men's participation is critical in achieving good maternal health outcomes.

Quality of written
I would caution the authors from making a statement that men are the 'primary stakeholders' or 'gatekeepers' in women's health. While you make the point that is certainly true in many societies, it important to keep in mind that how women access care is more complex. The root causes which are economic and social go beyond men.
A minor point about semantics: I would avoid using PIH-either use gestational hypertension (or pre-eclampsia-I could not tell from the questions what you were targeting as these two entities are very different while being on the same spectrum). Secondly, instead of 'developing countries', I would suggest 'low income countries'. I am concerned that the language and the level of knowledge required to understand the questionnaire is indeed very high. You do rightfully point this out. Can you explain why you chose such a sophisticated level? How does this relate back to your objective of assessing knowledge around morbidity? The questions are at the level a health care professional might answer. A small percentage of your sample didn't even have any education! Most questionnaires should be targeted to a grade 6 (by North American standards) level.
F1000Research didn't even have any education! Most questionnaires should be targeted to a grade 6 (by North American standards) level.
From a public health perspective, the key messages should be to recognize that a woman is sick (perhaps even have a certain condition in mind) and to know what actions should be taken to prevent and/or treat the condition. Most certainly, the key is to seek care as soon as illness is recognized. I don't think the questions reflect this.
Secondly, some of the choices are wrong-for example, anemia does not present with severe headache. In fact, many serious conditions in pregnancy present with headache and this gives the impression that a less severe condition like anemia can present like this. Heel edema/ankle edema may be present in normal pregnant women and therefore, it is not a good choice. Bed rest is actually not recommended for the hypertensive disorders of pregnancy.
It would be interesting to know how men whose partners previously had these conditions scored compared to men whose partners did not have these conditions. I think the authors need to revise the discussion section with more limitations and as well propose how this knowledge can be used.