With technological advances and developments, internet access continues to increase [1]. Globally, approximately 4.53 billion people have access to web portals [2,3], with more than half using mobile devices and 38.5% using desktop computers to access the internet worldwide [4]. Increasingly, internet users are making use of the availability of web-based resources, with approximately 4.5% of all internet searches looking for health-related information [5-7]. Recently, nationwide lockdowns, social distancing, and restrictions due to the COVID-19 pandemic have led to an inevitable surge in internet use among individuals, including parents of young children, to seek health information on the web [8,9] and by health care practitioners to assist in service delivery [10,11]. Given that the internet offers considerable opportunities for immediate and easy access to web-based resources, it has become a significant medium for the dissemination of health-related information.

A universal growing demand for, and use of, web-based resources related to child health information is evident [12-15]. The consolidated behavior of using web-based resources for health information is particularly common among new and expecting parents, where they most frequently search the internet for information related to infant nutrition, development, social support, and health symptoms [16-22]. A 2016 Australian survey found that more than 73% of parents with children aged <5 years used websites and web-based forums to access child health–related information [23]. Interestingly, 30% of those parents reported not trusting the information sources [23]. Another recent study in Switzerland showed that 91% of parents with at least one child aged <2 years used the internet to search for information related to their child’s health and development [24]. The most frequently used sources reported were search engines (55%) and websites for parents (47%) [24]. Although the majority used the internet to search for health-related information, a large percentage of parents were skeptical about the trustworthiness of the web-based resources and their ability to correctly interpret the reliability of the health information they found [24]. This highlights the need for, and importance of, accessible websites to present health information accurately while ensuring it can be easily understood by their intended users.

This study updates and expands on a 2015 systematic assessment of infant feeding websites and mobile apps available in Australia [25]. The update of apps has been recently conducted by Cheng et al [26] in 2020. Hence, this study focused on updating and expanding the assessment of websites globally. The 2015 assessment found that 61% of Australian websites were of poor quality, with minimal coverage of infant feeding topics, lack of author credibility, and abstruse readability of content [25]. Since the publication of the 2015 systematic assessment, several other website assessments reviewing information related to infant health behaviors have identified similar findings [27,28]. A Korean study reported that websites were scored poorly when evaluated for availability, quality, and reliability of infant health information on the web [27]. Moreover, a recent analysis of 197 websites addressing preterm infants’ health information also found that the overall quality of websites was low to moderate in terms of reliability and content [28]. Provision of inadequate or incomplete infant health information on the web could result in parental confusion, apprehension, and poorer care for infants when parents are unable to evaluate the accuracy and credibility of the web-based information.

Over the past 10 years, websites have evolved from static read only to a more interactive and fully immersive experience [29,30]. Since the 2015 assessment of websites [25], there has been a marked growth in bandwidth levels enabling the design of more sophisticated websites to offer consumer-oriented health information in various interactive ways, such as videos, parent forums, podcasts, and multilingual options that have provided a context to support culturally diverse people across the world [31-38]. In addition, the emergence of artificial intelligence and machine learning since 2016 has given rise to chatbot technology that simulates human-like conversations to provide consumers with support and relevant information [39]. This has been tested and proven to be successful among parents of young infants searching for information related to infant sleep and feeding practices [40].

With more parents resorting to web-based sources to seek infant health information, it is imperative for resources on the internet to reflect the latest infant and child health guidelines. Therefore, the aim of this study was to update the 2015 systematic assessment of websites [25] by evaluating the content, suitability, readability, comprehensibility, and quality of information targeting infant nutrition, active play, screen time, and sleep behaviors on websites globally. In addition, this review expanded the 2015 systematic assessment [25] by examining interactivity, features, and cultural considerations of the websites.

A systematic search and assessment were conducted to identify and evaluate websites targeting infant feeding, active play, screen time, and sleep behaviors between July 2021 and February 2022. As shown in Table 1, a range of validated tools was used to assess the selected websites. Details of the evaluation tools are described in Multimedia Appendix 1, and the details of the methods are given in Multimedia Appendix 2.

As shown in Figure 1, a total of 450 global websites were reviewed between August 2021 and February 2022. The removal of 218 duplicate websites left a total of 232 unique websites. Of these, 66 websites met the inclusion criteria and were eligible to be evaluated. The remaining 166 websites were excluded, as 50 were not relevant to infant health behaviors; 38 were articles, PDF documents, or downloadable documents; 42 were a web-based shops; 23 had insufficient content; 4 were government guidelines; 7 offered web-based consultations, and 2 were infant-related apps on Google Play and App Store.

Overall, 3% (2/66 assessments) of websites were rated superior for suitability of health information, 82% (54/66 assessments) of websites as adequate, and 15% (10/66 assessments) of websites not suitable, as shown in Table 3 (Multimedia Appendix 3).

Very few websites were rated superior on literacy demand, such as writing style, context, and vocabulary used. Overall, 15% of websites provided culturally appropriate visual aids based on the consumers they were targeting. There were variations in the type of images used in the resources. Some images depicted different sex, race, color, religion, and age, whereas others targeted specific cultural groups such as Indian and Aboriginal and Torres Strait Islander. Only one of the sites addressed the cultural specificity of information relating to experience, language, or provision of examples to patients from diverse sociodemographic backgrounds.

Overall, 21% (14/66) of websites had the option to be translated into a language other than English, such as Arabic, Spanish, Hindi, or Bengali. Only 14% (9/66) presented information that addressed culture in texts or images. The culturally appropriate information varied between fasting tips and breastfeeding in Islam, Christianity or Judaism, fasting and pregnancy tips, multiracial graphics, and recipes.

Many websites provided a clear layout of information and easily understandable cover graphics that clearly portrayed the purpose of the material. Most topics were subdivided to improve readers’ self-efficacy, which was rated as adequate; for instance, infant sleep was subdivided into quiet playtime, bedtime routine, safety sleep practices, and infant sleep recommendations.

Very few websites met the Australian Federal government’s recommended level for written health information of lower than a grade 8 reading level: 29% (19/66) of websites (SMOG), 20% (13/66) of websites (F-K web-based tool), and 12% (8/66) of websites (consensus tool). In 2 of the websites assessed for interreliability, the readability scores ranged from 8 to 11 and 7 to 14 between the researchers depending on the varying content selected for assessment (Multimedia Appendix 4).

The median readability grades were 8.5 (IQR 7-10), 9 (IQR 8-11), and 10 (IQR 8-11) using the SMOG formula, web-based F-K calculator, and consensus calculator, respectively. There was a good correlation among reading grade scores across the readability measures (P<.001; 2-tailed).

Table 4 presents the results of the websites interactivity scores in terms of active control, two-way communication, and synchronicity.

The overall interactivity of websites was predominantly poor (49/66, 74%). The remaining 26% (17/66) of websites were marginally adequate with no websites obtaining an excellent rating. More than half the websites acquired an incomplete score for active control resulting from slow loading web pages and the inability of site search engines to return relevant results effectively. Very few websites encouraged visitors to talk back or facilitated 2-way communication between the visitors and the site. As for synchronicity of the sites, approximately one-fourth of websites received a full score (agree) for their ability to process input and respond to requests promptly (Multimedia Appendix 4).

The most common features found on websites were social media links (61/66, 92%), frequently asked questions (48/66, 73%), videos (44/66, 67%), and recipes (35/66, 53%), whereas language options, webinars, question and answer forums, chatbots, read out loud function, slide shows, animation, and games were less common. Moreover, 80% (53/66) of websites had additional features such as text font size options, tools (eg, ovulation calculator, pregnancy calculator, and parenting tools), download and print page content option, and YouTube accounts. Overall, 47% (31/66) of websites had associated apps on Google Play and Apple Store. Log-in options for personalized health information were presented on 40% (26/66) of the websites.

In this review, we systematically assessed 66 websites that reported health information related to infant nutrition, active play, screen time, or sleep behaviors. This review extends on the existing 2015 assessment by providing 2 main conceptual contributions. First, it covers the quality, content, suitability, readability, and comprehensibility of web-based infant health information at a global level. Second, it assesses the interactivity, features, and cultural considerations of webpages. In this section, we discuss the principal findings, comparison to prior work, implications for future practice, and then outline the strengths and limitations of this review.

This study found that the information content of the websites was overall poor in terms of scope and depth of information, which was similar to the findings of the previous 2015 assessment. Approximately one-third of the websites reported different advice on storage of expressed breast milk; for instance, “freshly expressed breast milk can be stored safely in the refrigerator for up to five days.” This information was contrary to the Australian guidelines on infant feeding that stated storage of expressed breast milk should not exceed 72 hours in the fridge [42]. This was due to the development of some websites in other countries, such as America or Europe, where guidelines differ from that of Australia [58,59].

The quality of websites was generally adequate when evaluated by HRWEF and mostly ranged from poor to good when rated by the QCSS tool. These findings are consistent with those from previous studies [60-63], which evaluated a range of health information available on the web using similar tools. This study also highlights that the quality of websites in terms of ownership, authorship, author qualification, purpose, referencing statements, and currency of information was the highest among media, nongovernmental organizations, and hospital websites and the lowest among university websites. This is an important finding, given that parents view university sites as a high-quality, reliable, and credible source of information [64]. It is vital that the health information on websites is continuously updated to meet the latest guidelines with relevant currency, authorship, qualification, and supporting attribution statements. This in turn will provide readers with the clarity they need to assess the quality of web-based health information and identify reputable websites. Moreover, websites with outdated information and no supporting statements can mislead readers, resulting in adverse health consequences [65-67].

Using the SAM tool, we found that website information on infant health behaviors was generally adequate. This finding is consistent with that of other studies on the SAM [68-70]. Despite the overall adequate suitability ratings of information on the selected websites, there were marked limitations in terms of cultural appropriateness, literacy demand, and illustrations. This highlights the issue that web-based infant health information rarely considered the needs of people from non–English-speaking ethnic groups and how they may interpret or apply the health information. This is unfortunate, given the ubiquitous nature of the internet. Given the increase in cultural diversity within Australia and abroad, it is important to consider cultural appropriateness of information and provide culturally and ethnically diverse consumers the capacity to access, understand, and use health information to make well-informed health decisions [71-73].

In regard to readability, this study highlighted that most websites were at readability levels beyond the ideal level of lower than grade 8. This finding was also reflected in previous studies [74-76]. Notably, a difference of 4 to 6 grades was observed between the interreliability readability grades of 2 websites scored by the 2 researchers DJ and HC. This reflected the inconsistent readability levels across various webpages within a website. Readability and health literacy play an integral part in information accessibility and usability [77]. The readability formulas are based on the number of words, sentence length, and number of syllables per word. Therefore, using simpler words, shorter sentences, pictures, videos, and co-design methods as per the Australian Commission for Safety and Quality in health care are important considerations for writing health information for consumers [78].

Furthermore, readability levels of web-based health information should be tested for consistency and presented in an easy-to-read format providing access to people with low health literacy.

The overall interactivity of website functions was poor. In addition, interactive features were mostly common among media and commercial websites and least common among government websites. According to the World Health Organization, various provision methods of providing health information are important to increase accessibility and achieve positive health outcomes [79]. Multiple patient-focused interventions have reported that various health information formats, such as videos, audios, and infographics, have contributed to the improvement in parental knowledge, satisfaction, and health outcomes [80-82]. Hence, it is crucial for credible websites such as government owned to make the wealth of information available on the web interactive and accessible to increase consumer engagement and use of reliable sources.

In comparison with the 2015 review, 55 new websites were assessed in this review, whereas 11 were common across both studies. It is important to note that the first 30 websites generated from every search term used resulted in a range of global websites for consumers to access. Hence, it is vital for education to be provided to parents of young children and health professionals that will enable them to determine the quality and credibility of web-based health information as accuracy is critical, especially in the first 1000 days of life [83,84].

Furthermore, several websites from the 2015 review were excluded for reasons such as their web-based content had been removed or they no longer exist. A potential reason may be due to the cost and maintenance of websites that were developed with limited funding. A systematic review of factors that influence eHealth reported that ongoing maintenance costs were barriers for several studies [85]. Another reason may be due to the evolution of websites over time and the inability of static websites to enable and host new features [30].

To replicate and compare results from the 2015 assessment, the same validated tools were used to assess the quality, suitability, and readability of infant health websites. Interestingly, the results of the new eligible websites did not greatly differ from the results of the 2015 systematic assessment [25]. There was a slight improvement in the quality of websites from poor to adequate or good measured by HRWEF and QCSS since 2015. However, the suitability rating of websites in both assessments measured by SAM was the same. Similarly, the readability of written health information in the majority of the websites did not meet government recommendation in both assessments as well. Although websites have evolved since 2015, minimal to no improvements have been identified in this review. One reason could be due to the lack of user involvement in the website design. A recent systematic review reported that a considerable number of studies raised concerns that involving users in technology design can be fairly demanding and requires time and effort [86]. Another reason may be the lack of use of validated tools such as the ones used in this study to ensure optimized quality during the development of websites.

Identifying cultural considerations, interactivity, and features of web pages is a value added to this study. We found that very few websites addressed culture or had interactive features such as multilingual options, chatbots, or read out loud functions. With the rise in immigrants from culturally and linguistically diverse communities [87], more and more people are facing access barriers to health information and eHealth services due to the lack of language support and culturally appropriate health information through the internet [88-90]. This demonstrates the need to engage end users throughout the web development to ensure high-quality outcomes and meet consumers’ needs and expectations.

With the wide spread of internet use and wealth of information available on the web [91], it is imperative for parents and health professionals to be guided to the optimal and most accurate sources of information. Websites should be screened for authorship, ownership, information date, and HONcode certification before use. There is a need for web developers to bear in mind their end users. One way to overcome this challenge is by involving consumers in website development through co-design workshops [92]. This will ensure developers get a good understanding of end users’ access requirements, literacy demands, preference for alternative presentation formats of information, and cultural considerations. An establishment of a regulatory body is also recommended to ensure that newly developed websites are built on validated tools and that all websites comply with government guidelines standards.

To the best of our knowledge, this is the first study to evaluate website interactivity, features, and cultural considerations for web-based infant health information. This review provides a comprehensive global overview of the available web-based information about infant health behaviors and identifies ways for improvement.

Although this study adhered to a rigorous systematic search process, there were several limitations. First, most websites included in this evaluation differed from that which was evaluated in the 2015 review. Hence, this demonstrates the dynamic nature and constant change of the internet. Thus, the website search during this review reflected a period that could potentially change. In addition, the website search was conducted in the English language using Google. Although Google is a highly used search engine, we acknowledge that some international users have access restraints [93]. Therefore, the results may not have identified websites present on other search engines or written in other languages. Furthermore, deleting cookies and search histories was intended to reduce unknown bias in the search strategy. However, it is acknowledged that the likelihood of most web users doing this is unlikely, and their searches might identify sites of which we were unaware.

Another limitation is that the assessment criteria used Australian guidelines. Therefore, there was a potential for websites following non-Australian guidelines to obtain an incorrect score on a few subtopics. Moreover, the interactivity scale used was originally meant to capture consumers’ perceptual assessment of websites. However, due to the lack of published validated tools used to measure interactivity of websites, the interactivity scale was adopted for the purpose of this study. Thus, the interactivity scale used was a subjective measurement based on the researcher’s experience on the websites. Nevertheless, the tools used offer a standardized way to best capture the quality and interactivity of web-based information.

As more parents seek web-based guidance on infant health behaviors globally, there remains a significant concern on the quality, readability, interactivity, and accessibility of websites promoting health behaviors during infancy. This systematic assessment revealed that there is a need for researchers and health care providers to leverage innovative web-based platforms to provide culturally responsive evidence-based information accessible to those with limited English proficiency. Furthermore, a focus is needed on continuously updating existing health websites in addition to recommending an establishment of a regulatory body to ensure compliance with government standards. Moreover, the development of new eHealth technology should be based on validated tools to ensure the optimal quality of websites.