We used Google, the largest online search engine, to query the first 20 results for each of the following 10 search terms: “lipoprotein(a),” “lipoprotein(a) cardiovascular risk,” “lipoprotein(a) elevated,” “lipoprotein(a) high,” “lipoprotein(a) levels,” “lipoprotein(a) screening,” “lipoprotein(a) test,” “Lp little a,” “Lp(a),” and “Lp(a) screening” [18]. Location, cookies, and user account information were disabled beforehand to avoid search bias. All 200 websites were accessed and downloaded as PDFs on November 5 to 6, 2020. OPEMs were defined as materials intended for patients and the public. Two independent reviewers reviewed each source’s mission statement or informational page to determine whether it was patient-facing. Research journal articles, advertised results, results failing to contain material on Lp(a), and non–patient-directed sources such as those intended only for health professionals or researchers were excluded.

This research is exempt from human participant institutional review board approval as no human participant data were used, and only publicly available OPEMs were included in the analysis.

OPEMs were grouped into 6 categories by primary source. Industry OPEMs were published by for-profit companies or offered a proprietary test or service related to Lp(a). Lay press OPEMs were published by news organizations and health care reporters that do not have a specific research or scientific focus. Research foundations and nonprofit OPEMs were published by health-related organizations that have a specific research or scientific focus. University or government OPEMs were published by academic institutions or national, state, or local government agencies. Clinic OPEMs were published by Williams Integracare clinic, an independently owned family medicine, chiropractic, and physical therapy clinic in St. Cloud, MN. Other includes Wikipedia.org, a crowdsourced online free encyclopedia, which published an article on Lp(a).

Websites meeting the criteria for OPEMs were converted into plain text in separate Word (Microsoft Corporation) documents and prepared in accordance with recommendations from the Centers for Medicare and Medicaid Services prior to scoring readability [19]. Using methods consistent with prior readability assessment studies we removed advertisements, videos, images, figures, captions, hyperlinks, disclaimers, copyright notices, acknowledgments, and citations. Periods were used to denote the end of all sentences; all other punctuation were removed. Symbols and numerals were spelled out to avoid artificial increases in reading level. Readability was assessed with Readable.com, as done in prior literature [16].

Similar to prior readability studies, grade-level readability was calculated using 5 standard readability metrics (automated readability index, SMOG index, Coleman-Liau index, Gunning Fog score, Flesch-Kincaid score) to produce robust point (mean) and interval (CI) estimates of readability. Averaging across multiple readability metrics has been demonstrated to yield more reliable results than relying on a single readability metric [20]. This was done to minimize bias because no single readability metric has been established as a “gold standard”; each readability metric is calculated differently and varies in limitations [21]. Generalized estimating equations were used to model grade-level readability by each search term, with the 5 readability scores nested within each OPEM [16]. Readability analyses were conducted with SAS Software 9.4 (SAS Institute), with sandwich estimation and the GLIMMIX procedure. All interval estimates were calculated for 95% confidence. The interval estimates reflect the variability of readability for the 5 readability metrics for each OPEM.

We found that the average reading grade level of OPEMs pertaining to Lp(a) generally exceeded AMA readability recommendations that OPEMs be written at or below a sixth grade reading level to be accessible to the public. Average grade-level readability of OPEMs on Lp(a) was a 12.2 (95% CI 10.9798-13.3978) grade level, exceeding the average reading level of US adults (eighth grade) [13]. Of 27 unique websites reviewed, only 1 website had a lower bound of reading grade level (5.7) that was below the sixth grade reading level recommendation. Our results suggest that the overwhelming majority of Lp(a) OPEMs are written at a reading level that is too high for the minimally health literate members of the public.

These findings have several important implications for how patients may make decisions about this important, proatherogenic lipid fraction that is receiving increased attention. Patients frequently use the internet to supplement health information from their clinicians [32]. If presented clearly, online health information can be a valuable patient resource. Patients who feel more informed are more comfortable asking their provider questions and report better understanding of their providers’ explanations and greater self-confidence in making health care decisions [33]. A 2018 cross-sectional survey found that patients who searched online for health information to solve their medical problems were also significantly more likely to change their medical decision based on information gathered [34]. It is noteworthy that OPEMs in the academic and government categories were the most readable. Academic and government sites are regarded as reliable sources given that many organizations seek to advance public health and knowledge [35]. Given the major influence of online health information on decision-making, these academic and government websites may also benefit from direct guidance to craft OPEMs with readability targets in view. Despite the importance of these OPEMs in reaching the lay community, research foundations and nonprofit sites demonstrate the largest gaps between patient reading skills and OPEMs reading level on Lp(a).

The less readable OPEMs tend to cover topics in greater depth, including nuances around Lp(a) measurement and standardization, and the role of Lp(a) in thrombogenesis and wound healing. Thus, there appears to be a trade-off between readability and comprehensiveness in OPEMs.

These findings align with results from prior studies across a broad range of health conditions that show OPEMs commonly exceed the recommended readability level. Ayyaswami et al [36] showed that greater than 99% of OPEMs relating to cardiovascular disease were written above the grade level recommended by the AMA [35]. OPEMs are frequently written at a reading level too difficult for the public to comprehend, and low readability levels have been documented across disciplines including common topics related to surgery, oncology, and radiology [37,38].

The percentage of adults with below basic health literacy is considerably higher for populations who identify as Black (24%), Hispanic (41%), American Indian/Alaska Native (25%), and Asian/Pacific Islander (13%) compared to non-Hispanic White (9%) [39]. Similarly, over half of adults older than 65 years were found to have less than a basic health literacy level [39]. These are the very populations known to face a disproportionate burden of cardiovascular risk. Our current risk prediction models such as the Pooled Cohort Equations do not adequately capture the risk to heterogenous racial/ethnic groups [1,40-42]. The importance of reaching these populations is further increased given that Lp(a) is known to circulate at higher levels in patients of African and South Asian descent [43,44].

Many clinical preventive and screening services are underused by historically marginalized racial/ethnic communities and older adults due to inadequate health care access and low health literacy, but we suggest that providing more readable OPEMs may help bridge this gap in care. In the interim, our study findings remind clinicians to consider the readability of OPEMs and patient literacy when recommending Lp(a) evaluation. Shared decision-making requires adequate understanding of the risks and benefits of any diagnostic testing or risk stratification procedure. Actionability of OPEMs on Lp(a) is presently limited by the lack of approved therapies; however, with changes in emerging therapies and practice guidelines, understandability and actionability will be important parameters to assess systematically using the Patient Education Materials Assessment Tool in future studies.

A strength of our study is that we incorporated readability results from five different standard readability metrics, which allows us to have a robust evaluation of readability regardless of the number of websites evaluated. All of the reading grade level estimates supported our hypothesis that Lp(a) OPEMs are written above the recommended sixth grade reading level. We also included a thorough review of possible patient queries by analyzing 200 search results for 10 commonly used search terms. Finally, we included examples of communication of a concept from both more readable and less readable OPEMs. This may serve as a real-world, practical guide for creators of OPEMs seeking to choose words and phrases that will be accessible to a broad audience.

Our study should be interpreted in the context of certain limitations. We did not account for other search engines besides Google. This limitation is somewhat mitigated by the fact that 88% of global internet users use Google as their most frequent search engine [18]. As with other OPEM readability studies, the readability metrics used here do not consider the inherent complexity of some medical terms. Polysyllabic words and longer words are automatically rated as more complex and less understandable than short or monosyllabic words, which in medicine does not always hold true.

In conclusion, we found that the grade level readability of OPEMs relating to Lp(a) generally substantially exceeded the sixth grade reading level recommended by the AMA. This gap in readability may disproportionately affect patients with low health literacy. Creators of OPEMs should be mindful of the readability of their content. Ensuring that online content is understandable by broad audiences is a necessary component of increasing the impact of novel therapeutics and recommendations regarding Lp(a).