In October 2013, the following key terms were entered separately in the two most commonly used search engines, Bing and Google [18]: “Congenital Heart Disease”, “Congenital Heart Defect”, “Ultrasound Heart Disease”, “Ultrasound Heart Defect”, “Pregnancy Heart Disease”, and “Pregnancy Heart Defect”. The inclusion criterion was a website written in English that provided patient information regarding CHD. The search was made in incognito mode in order to minimize influence from previous search patterns. The generated result of the search procedure was saved, and the first 50 hits obtained for each search procedure were screened for inclusion (N=600). Duplicate websites and direct links to communities/blogs, video materials, and scientific articles were excluded. In total, 533 (88.8%) were excluded, leaving 67 (11.2%) websites for inclusion in the study. Figure 1 presents the selection process, that is, key terms, hits, and excluded and included websites via Bing and Google, respectively.

The websites we included were affiliated with independent information websites (25/67, 37%), charity/private organizations (17/67, 25%), hospitals/clinics (17/67, 25%), governments (4/67, 6%), medical companies (1/67, 2%), and other websites (4/67, 6%). The majority of the websites could not be attributed to an author (51/67, 76%), whereas a minority could be attributed to medical professionals (10/67, 15%), journalists (2/67, 3%), and others (4/67, 6%).

Central subjects of content on the 67 websites were categorized into 25 categories with 46 subcategories (Table 1).

The majority (>70%) of the websites contained information about care and treatment of children with CHD (88%, 59/67), causes of CHD (88%, 59/67), symptoms of CHD (85%, 57/67), prevalence of CHD (81%, 54/67), potential complications of CHD (75%, 50/67), prenatal diagnostics/screening methods (72%, 48/67), and specific CHD (72%, 48/67). A minority (<30%) of the websites contained information about common feelings following prenatal diagnosis of CHD (27%, 18/67), postnatal quality of life (27%, 18/67), examples of previous cases that continued the pregnancy (24%, 16/67), precision of prenatal diagnosis of CHD (24%, 16/67), delivery (21%, 14/67), postnatal coping with the diagnosis (19%, 13/67), risks of CHD in future pregnancy (15%, 10/67), fetal intervention (12%, 8/67), presentation of the multidisciplinary team in care of the child (12%, 8/67), termination of pregnancy (6%, 4/67), care during pregnancy (5%, 3/67), and information specifically directed to partners (1%, 1/67).

We searched the Internet with different key terms to find publicly available patient information following a prenatal diagnosis of a congenital heart defect. Hits ranged from 350,000 to 67,500,000 and 67 of 600 screened websites were included in the study. Over a third (37%, 25/67) of the websites were affiliated with independent information sources, whereas a quarter (25%, 17/67) were affiliated with hospitals/clinics. The majority of the information on the websites (76%, 51/67) could not be attributed to an author. A minority of the websites contained information regarding certain prenatal aspects, that is, common feelings following prenatal diagnosis of CHD, precision of prenatal diagnosis, delivery, risks of CHD in future pregnancy, fetal intervention, termination of pregnancy, and care during pregnancy. Furthermore, the majority of the included websites were scored very poor by the DISCERN instrument with regard to information about effects of no treatment, support for shared decision making, achievement of its aims, explicit aims, risks of each treatment, how treatment choices affect overall quality of life, and areas of uncertainty. The reliability of the included websites was poor according to both the DISCERN instrument and the LIDA tool, particularly regarding currency, content production, aims of website, and areas of uncertainty. However, the accessibility and usability of the included websites were sufficient.

The literature suggests that the overwhelming number of websites found when searching the Web for information leads to information overload and searching difficulties [5,12,24]. The fact that it is difficult to find relevant information on the Internet is exemplified in this study: 39.3% (236/600) of the screened websites were irrelevant. Thus, it is possible that persons seeking information about CHD miss accurate and valuable information or give up information retrieval via the Internet because of difficulties in finding relevant sources. This difficulty might be enhanced by the fact that cardiologists seldom give recommendations on websites in connection with the diagnosis [17]. Health care professionals need to be aware that parents of children with CHD rank information regarding websites at the time of diagnosis as more important than cardiologists [25]. Consequently, as health care consumers are increasingly using the Internet to search for information [1-4], health care professionals need to address these circumstances and actively strive to recommend and provide accurate and reliable high-quality information online.

It seems that the websites target families following a postnatal diagnosis or women opting to continue the pregnancy. Previous research suggests that induced abortion is viewed as a socially unacceptable and stigmatizing procedure [26], independent of state laws on pregnancy termination [27]. It could be speculated that this perspective influenced the content of the included websites, as few contained information regarding termination as an option following a prenatal diagnosis of CHD.

The Internet may provide inaccurate and biased material [5,28,29]. It is imperative that pregnant women make informed decisions regarding whether to continue or terminate the pregnancy, which may be hindered by unreliable information sources found online. The majority of the websites in this study had poor reliability in a number of areas, including currency, conflicts of interest, and content production. The importance of current and unbiased information is especially important in the context of the rapidly expanding and evolving field of fetal cardiology, in order to promote informed decisions.

Health literacy, that is, the degree to which individuals have the capacity to obtain, process, and understand the health information and services needed to make appropriate decisions [30], is an important concept when discussing disparities in health information comprehension. Illustrations as a complement to oral information can substantially increase comprehension of health information, are especially helpful for those with poor health literacy [31], and are desired in connection with initial diagnosis [12]. However, animations and illustrations were scarce among the reviewed Web pages. It is therefore possible that the information online is not suited for those with poor health literacy. Health care professionals need to acknowledge this and provide pedagogic tools to promote patient comprehension and equal care.

This study did not evaluate the scientific quality of the reviewed websites, nor did it assess the accuracy of the information found, that is, if the included websites contained any inaccurate or misleading information. Furthermore, it is possible that the key terms do not fully represent the online landscape of websites about CHD and that other results would have emerged with different search methods. However, according to previous research, the majority of health information seekers use search engines as their primary source [1,32], and the search engines used in this study are at present the most commonly used [18]. Moreover, the searches yielded 136/600 (22.7%) duplicate websites, indicating saturation and that the searches do represent the online landscape.

The DISCERN instrument and LIDA tool are based on subjective ratings. Only the first author conducted the assessments, and this could imply poor generalizability, and perhaps also a certain bias. The DISCERN instrument, developed and designed to help users of consumer health information judge the quality of written information, has been found to be consistently understood as well as transferable to different specialties (eg, [20,23,33,34]). Furthermore, the first author is a nurse, which could possibly indicate different views than non-professionals and thus different scorings. However, previous research suggests that scorings are not dependent on previous knowledge of the specific condition [22], and it has been concluded that health professionals score DISCERN similarly to non-professionals when assessing health information [33]. Taken together, we find it reasonable to assume that the main outcomes from this study would have been similar even with another evaluator or with more than one evaluator.

Approximately 25% (17/67) of the websites were affiliated with hospital/clinics, and 76% (51/67) could not be attributed to an author. Consequently, it is important to bear in mind that the information found on the included websites may differ from the information provided by health care professionals following a diagnosis of CHD.

It remains unclear if websites about congenital heart defects following a prenatal diagnosis contain accurate and suitable information. This needs to be evaluated in future studies by health professionals within fetal/pediatric cardiology and persons with experience of a prenatal diagnosis of a CHD.

It has been reported that expectant parents want more information than that provided by health care professionals following a prenatal diagnosis of CHD [17]. The results from this study, however, indicate that existing websites do not adequately supplement counseling. Development of an information source via the Internet would enable expectant parents to access accurate and tailored information that complements the standard counseling offered today. In order to evaluate such tools, national, or even international, research collaborations are needed.

Easily accessible information on the Internet influences the conditions for doctor-patient interaction [35]. What can be communicated by the doctor, and acknowledged by the patient, always depends on the previous knowledge and perspective that the patient has developed in their own information seeking. Thus, more knowledge is needed on the communication chains in which the patients build their knowledge and understanding.

The linguistic readability of the websites was not assessed in this study. Neither was patients’ interpretation and evaluation of the information investigated. Several models for mechanical syntactic analysis, in order to measure readability, were developed in the early years of text linguistics, focusing mainly on factors such as word length and syntactic complexity [36,37]. Similar models have been used to assess medical information [23,34]. Later research in computational linguistics suggests that measures of semantics and discourse cohesion, that is, “content”, show higher correlations with reported readability [38]. To gain a deeper understanding of Web texts, high-quality qualitative text analytical studies are needed. Preferably, such studies should be combined with reader interviews, focusing on interpretation and comprehension. Furthermore, additional research is needed in order to understand how different types of illustrations can provide relevant understanding of the disease.

The reviewed websites do not adequately supplement counseling from health care professionals following a prenatal diagnosis of CHD. Difficulties in finding relevant information sources using Web search engines and quality deficits on websites are an incentive for health professionals to take an active part in providing adequate and reliable information online about CHD. Future websites need to have a clearer prenatal perspective to become a source of knowledge for prospective parents seeking information online following diagnosis.