In this study, following the Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) guideline [17], we performed a systematic review of the literature regarding pregnancy-related consumer information needs from 2009 to 2019 from 4 major databases: CINAHL, Cochrane Reviews, PubMed, and Web of Science using the search queries “health information” and (pregnan* or fertility or infertility or conception or mother* or matern* or prenat* or pre-nat* or antenat* or ante-nat* or perinat* or peri-nat* or pre-pregnancy or pre-pregna* or gestation*).

In total, we found 4583 articles. After removing duplicates, 2712 articles remained. We excluded 2573 articles after the title and abstract screening. Then we performed full-text review on the remaining 139 articles, including articles about pregnancy information needs or sources, pregnancy- or infertility-related, that were published in English from 2009 to 2019 and excluding articles that were not full papers, were opinion papers, or did not contain as abstract. The remaining 33 articles were evaluated using the Critical Appraisal Skills Program (CASP) [18] for quality assurance and were determined to have adequate quality to be included in the final full-text extraction. The detailed information about CASP review can be found in the Multimedia Appendix 1. The PRISMA workflow is shown in Figure 1.

In the evaluation of the full text of the 33 articles, we identified the following aspects of information: (1) year when the article was published, (2) topic of the study, (3) research method of the study, (4) sample size of the study (if human subjects were used), (5) sources of the health information, (6) stages of pregnancy on which the study focused, (7) target population, (8) country, and (9) consumer information needs.

The trend of the number of the included articles is shown in Figure 2. Between 2009 and 2015, there was a small number of studies about pregnancy-related consumer information needs and sources. Since 2015, there has been a surge in the number of studies published.

Two articles pointed out that consumers face certain barriers to accessing pregnancy-related information [13,33]. Kamali et al [33] found that lack of knowledge and ignorance regarding existing resources are 2 frequent barriers to accessing health information among pregnant women. In addition, the systematic literature review by Ghiasi [13] classified the barriers to accessing health information into 7 categories: (1) affective barrier included feeling ashamed or embarrassed to talk about pregnancy-related issues; (2) cognitive barrier included negative attitudes of health care providers, lack of familiarity with the internet for information seeking, poor patient-provider communication, reliance on self, underestimation of the risks of pregnancy complications, and unawareness of relevant information sources; (3) cultural barrier mainly referred to the social stigma related to pregnancy; (4) availability barrier included long wait times to see a doctor in clinic, lack of adequate information sources, and inadequate information from health care providers; (5) accessibility barrier referred to lack of time; (6) affordability barrier included high cost of information sources; and (7) infrastructure barrier referred to erratic power supply to access electronic information.

In addition, as shown in the reviewed papers, there were 3 scenarios regarding the accessibility of information: (1) needed information does not exist [12,41], (2) needed information exists but is not easily accessible (eg, consumers do not know how to find it or it is not easily readable) [32,39,42], and (3) needed information exists but in an incomprehensive or inconsistent manner [12].

Several studies also assessed the quality of information provided by these sources. For example, Brown et al [29] investigated the quality of information provided by iPhone apps for pregnancy and nutrition using the Mobile Application Rating Scale tool; the Coventry, Aberdeen and London–Refined taxonomy; and expert review. The results showed that the included apps were generally of moderate quality. De Man et al [39] assessed the availability and quality of online health information about fertility in the cancer setting using the DISCERN instrument, the Minervation validation instrument for health care websites, and a readability test. The results suggested that the included websites had moderate scores in readability and usability. However, the readability test did not consider medical jargon. Carlsson et al [40] assessed the quality of online information related to congenital heart defect following a prenatal diagnosis. The results showed that the majority of included websites scored poor in quality and readability with respect to the various themes about congenital heart defects.

A secondary contribution of this study is an ontology about pregnancy information needs that contains information across all stages of pregnancy (prepregnancy, during pregnancy, and postpartum). To develop the preliminary version of Pregnancy Information Needs Ontology (PINO), one researcher (YL) first enumerated important terms about pregnancy information needs based on all reviewed papers. The goal of the enumeration was to obtain a comprehensive list of the terms. As certain terms may overlap with others in the meaning they represent, the researcher then synthesized terms that convey similar meanings. In the next step, the researcher developed a class hierarchy using a top-down approach, with the resulting hierarchy organizing information into themes and subnodes. The last step defined the properties of terms. PINO consists of 267 classes, 555 axioms, and 271 subclass relationships. In future work, we will evaluate both the intrinsic aspects (eg, concept orientation, consistency, soundness) and extrinsic aspects with the support of tools [43,44]. Table 1,2, 3, and 4 present subtaxonomies of PINO for each stage of the pregnancy trajectory, including prepregnancy, during pregnancy, postpartum, and across all stages.

Figure 3 presents the network graph for the co-occurrence of consumer information needs in all 33 articles. Purple, blue, green, and yellow nodes refer to prepregnancy, during pregnancy, postpartum, and multistage, respectively. Gray, orange, red, and black edges refer to the co-occurrence of 1, 2, 3, and 4 times, respectively. For example, consumer information needs related to infertility co-occur with medication in pregnancy, health care products, lab test, nutrition, physical exercise, or lifestyle in pregnancy, pregnancy taboo, prenatal care, labor/delivery, and maternal recovery once, respectively. Consumer information needs about labor/delivery co-occur with maternal recovery once, symptoms of pregnancy and diagnosis twice, test 3 times, and prenatal care 4 times. Consumer information needs about family planning co-occur with newborn care, postpartum care, and mental health once, respectively. Consumer information needs about mental health co-occur with health care products, insurance, oral health, parenting, planning for a newborn, and postpartum planning, respectively.

For all human-subject studies in this review, Figure 4 presents the network graph for intersections between consumer information needs and countries of participants. For example, participants in Ghana sought information related to confirming pregnancy, labor/delivery, nutrition, physical exercise, and lifestyle in pregnancy, while information about medication in pregnancy was sought by participants in Belgium, Canada, Iran, Netherlands, and the United States. This figure only represents the demographics of needs of participants by country in all reviewed articles and could not be generalized to describe the needs of consumers in each of these countries.

Pregnancy is a major life-changing event and is considered one of the most sensitive periods of a women’s life. Special care and various types of information are required during this period. However, due to reasons like health literacy and technology proficiency [32,37,38], women feel pregnancy-related information is insufficient and not well organized. Therefore, it is important to organize such information in a systematic way to meet the consumer information needs of this population. In this study, we assessed pregnancy-related consumer information needs and available information from different sources through the review of relevant articles. To the best of our knowledge, this is the first study to systematically assess consumers’ pregnancy-related consumer information needs across all stages of the pregnancy trajectory. It is also the first study to derive a taxonomy of pregnancy-related information through a systematic literature review. Existing taxonomies of pregnancy-related needs in the included articles mainly focused on one or two stages of pregnancy [10,13,21,22,24,27,30,33,41] or a facet of pregnancy (eg, female infertility [12], nutrition [28,29,42], physical activity [28,48], vaccines [34], medications [25], labor analgesia [47], congenital heart defects [40], infant feeding [36], or miscarriage [23]). Our study systematically assessed consumer information needs and sources across the entire pregnancy trajectory, including preconception, during pregnancy, and postpartum.

In the last 50 years, the role of the patient has changed. Previously medical professionals were seen as the sole decision makers and information providers. Now patients are more responsible for decision making and personal information seeking. Given the nature of pregnancy decisions, this has definitely impacted this population. Consumers’ online information seeking about pregnancy could be to obtain a sense of empowerment and preparedness in interacting with health care providers [49]. They also need to be able to understand the information presented by health care providers and better monitor decisions made with doctors [49]. High-quality information on the internet is often appreciated by consumers so that they can also use this information for decision making [50].

There are many theories that have been developed to provide a framework to examine information behaviors. One theory that applies to pregnancy information seeking is Dervin’s sense-making theory [51]. The goal of sense-making may not always be finding relevant information but may include finding people in similar situations and avoiding bad information. Sense-making is not new and not limited to the information fields. It has also been seen in organization, communication, education, and human-computer interaction. In health care, it has been used in diabetes management [52], tumor contouring [53], and Down syndrome diagnosis [54]. One premise of this theory is the connection between how someone looks at a situation and how they are able to understand or make sense of this situation. Think about the gap between what you know and what you need or want to know. In pregnancy, this gap is almost guaranteed even in individuals who have had successful pregnancies in the past. This gap can be influenced by many personal factors including history, previous education, adaptability, and skill at building connections with information. Ultimately, these factors are more individual and personal then situational. This moves the assumptions that are made from the individual to the instances of information seeking. This allows for studies like this to focus on the differences in the information and group the population together even though there are many differences in this population.

Traditional information sources (eg, health care providers) often provide critical and general information about pregnancy. However, such information is often not contextualized in personal situations. In addition, the medical jargon used in the information presented in traditional sources makes it difficult for consumers to understand [55]. As a result, the internet has become an alternative source of information. During the online information-seeking process, health literacy plays an important role. Prior studies found that consumers with low health literacy tend to use the internet less frequently [56,57] and find it challenging to retrieve and appraise online health information about pregnancy, which could in turn impair their ability in decision making and achieving better pregnancy outcomes [58]. Furthermore, our study found that only 2 out of the 33 articles about consumer information needs and sources assessed the barrier to accessing pregnancy-related information online and no article investigated the impact of health literacy on consumer information needs related to pregnancy. Therefore, future work should carefully examine consumers’ health literacy so that information can be provided to consumers based on their different levels of health literacy.

We also found that most studies in this review assessed consumers’ pregnancy-related information needs and the available information from different sources. However, only a few studies examined the quality of the information provided by these sources, which is consistent with the finding of the previous review [14]. Among the information quality criteria, readability was one of the most frequently cited issues [59]. Berland et al [7] found that English and Spanish websites required high levels of reading ability. Also, consumers often find that online health information contains a high level of technicality with lots of medical jargon [60]. To make better use of the internet, the readability of the content should be presented at or below a 5th grade reading level to accommodate people of all health literacy levels [61]. Last, the booming production of online health information has resulted in information overload [62]. To address these issues, it is important that health care providers and supportive technologies develop ways to direct consumers to high-quality sources that are layperson-friendly and pertinent to their situation.

In this review, we also intended to identify the types of information that are needed by consumers but have not been provided by different sources including websites, mobile apps, and health care providers. However, there are a few reasons why this goal could not be achieved. First, most of the articles are about consumer information needs but only a limited number of articles are about information sources. Also, the focus and ways of organizing information in these articles are different. For example, some articles focused on a specific stage of pregnancy while others focused on a specific disease or complication related to pregnancy, and thus there were not many common themes of information provided. Second, the included articles about information sources could not represent the comprehensive types of information available in different platforms. Hence, future work can consider using systematic approaches to assess themes of information about pregnancy available on all the different platforms.

In addition, our study resulted in a taxonomy of pregnancy-related information. This taxonomy can be implemented into supportive technologies (eg, webpages, smartphone apps, and patient portals) so that consumers can easily access and retrieve a structured body of information about pregnancy regardless of the stage of their pregnancy. We have uploaded our PINO taxonomy to our GitHub repository [63] and BioPortal [64]. Currently, PINO only includes 267 classes and 271 hierarchical relationships. In the future, we will improve PINO by adding more concepts and semantic relationships between concepts to further express the conceptual domain space across all stages of pregnancy—prepregnancy, during pregnancy, and postpartum.

The result of this review suggested several research gaps. First, only a few studies assessed information quality of the sources of pregnancy-related information. Furthermore, these studies suggested that the quality of the provided information is questionable. Given that prior work has proposed a model for assessing consumer health information quality [65], future work can systematically extract and assess the quality of pregnancy-related information from all available sources (eg, providers’ webpages, online health forums, mobile apps).

In addition, this review suggests that very little is known about how much consumer information needs about pregnancy have been satisfied and what needs are not yet met. Future work can consider comprehensively examining both consumer satisfaction and unmet consumer information needs about pregnancy.

Last, given the overwhelming amount of available health information online and the filter failure of existing information retrieval systems [62], future work could explore ways to help consumers retrieve high-quality, customized, and layperson-friendly health information.

This review has limitations. First, we only systematically assessed articles about pregnancy-related consumer information needs and sources published from 2009 to 2019. Second, articles in this study have different focuses and organization of pregnancy-related information themes. Neither consumer information needs nor available information could be generalized to all pregnant women or sources. Hence, conclusions regarding which consumer information needs have been met could not be made.

In this study, we reviewed 33 articles published from 2009 to 2019 about pregnancy-related consumer information needs and available information from different sources. The resulting taxonomy comprehensively covered and provided hierarchal themes of pregnancy-related consumer information needs across the stages of pregnancy. Last, findings of this study suggested several future research directions: systematically assessing the quality of pregnancy-related information from all available sources, comprehensively examining both consumer satisfaction and unmet consumer information needs about pregnancy, and exploring ways to help consumers retrieve high-quality, customized, and layperson-friendly health information.