Reframing healthy food choices: a content analysis of Australian healthy eating blogs

This study was conducted using a content analysis approach to identify characteristics of popular healthy eating blogs. Content analysis is a research method that has been increasingly used to analyse digital content, including written and visual content in internet forums, websites and in social media platforms [26‐29]. Australian healthy eating blogs authored by self-identified health professionals were identified and followed for three months (December 2017–March 2018).

A purposive and snowball sampling approach was used to identify healthy eating blogs from search engines including Google, Bing and Yahoo (Fig. 1). The search terms used in each search engine included ‘Australian Healthy Eating Blogs’ and ‘Top 100 Australian Healthy Eating Blogs’. There is no precise method to estimate the number of online healthy eating blogs [30, 31]. Therefore, blogs were identified by focusing on the first page of each search engine, as it was assumed they are the most widely read and influential [30‐32]. This decision was driven by a study conducted by Jansen and Spink (2009) that investigated consumer search engine click-through behaviour and reported 73% of consumers did not move beyond the first page of search engine results [32]. Additionally, current knowledge of search engine optimization (SEO) suggests that first page search engine results have higher website traffic [33].

Each webpage included was systematically searched and assessed for eligibility. Healthy Eating blogs were deemed eligible if: [1] the author self-identified as a health professional [2]; the blog was written by a single author [3]; the blog was written by an Australian author [4]; the blog had a minimum of one post per month, and [5] the blog focused on communicating healthy eating information to the general adult population. Blogs were excluded if they focused on a specific condition or disease state, if they could not be accessed due to a broken web link or if they were privately accessible. Additionally, micro-blogs which are defined as short posts restricted to 140 characters, were not included for analysis, as the primary focus of this study is to explore blog characteristics and the ways in which healthy eating messages are communicated [34, 35]. Due to the character restrictions, the intended purpose of micro-blogs is to provide brief updates [34, 35].

In this study, authors who identified as a health professional were chosen because they are already an accepted source of nutrition and health information. It was assumed that blogs written for the exclusive purpose of providing healthy eating information by health professionals would aim to provide evidence-based information, which may not be case for personal blogs. Health professionals in this study included, Accredited Practicing Dietitians (APDs), Nutritionists, Wellness and/or Health Coaches, Personal Trainers, General Practitioners, or any combination of the above. The qualifications of authors were self-reported and recorded on the ‘about me’ section of each blog; they were not further verified by the research team.

A total of 14 healthy eating blogs were identified for analysis. Each blog was routinely viewed over the three-month collection period to ensure blogging frequency met inclusion criteria (9 blogs did not meet criteria and were subsequently excluded). Additionally, due to the temporary nature of the Internet and to ensure no adaptations to posts occurred, screenshots were taken of each blog post and stored as separate Word Documents.

The study was approved by the relevant institutional Human Research Ethics Committee [removed for blind peer-review]. Conducting internet research presents challenges that need to be acknowledged, including the boundary between public and private content in relation to consent [36]. Hookway (2008) suggests that ‘accessible blogs may be personal but they are not private’ [37]. In this study, blogs were considered public if they were not privately accessible or password protected, therefore the institutional ethics committee did not require individual consent from blog authors. However, to preserve anonymity, blogs were de-identified for the purpose of the content analysis and reporting.

A coding scheme was derived from previous research on the content of information communicated within health websites [19, 20, 38, 39]. The Health-Related Website Evaluation Form (HRWEF) and the Suitability Assessment of Material (SAM) were adapted to create a coding scheme to help guide analysis [19, 20, 38‐41]. The HRWEF evaluates the content, credibility, currency, accuracy, reliability, readability and design of health-related websites, based on a scoring system [19, 20, 38, 39] . The SAM tool evaluates literacy, readability, writing style, vocabulary, context, graphics and illustrations, layout, typography, interaction and cultural appropriateness of health-related websites [19, 20]. Each variable was coded through a scoring system which reflected the degree to which they aligned with criteria adapted from the HRWEF and SAM. A priori content analysis was conducted on four blog posts from identified healthy eating blogs that were dated prior to the commencement of data collection to ensure applicability of the coding scheme and for coding training purposes.

Coding was conducted by two researchers [blinded for peer-review] using Microsoft Excel. Both researchers had formal University training within the field of Nutrition and Dietetics. Coding was undertaken collaboratively to aid reliability and consistency of the coding process. Each blog was assigned an identification letter and each post within a blog was assigned with an identification number. This allowed for the two researchers to discuss and compare coding during analysis. Differences in code assignments were resolved by the two researchers by discussing the rationale behind the coding decision and determining an agreed coding assignment. Additionally, in order to facilitate transparency, researchers documented their own thoughts about the purpose of the blog post at the completion of analysis which were then collaboratively shared between the researchers.

A total of five blogs were followed over a three-month period from which 90 blog posts were identified. From the 90 blog posts, 14 blog posts were excluded due to content being non-nutrition related, resulting in a total of 76 blog posts for analysis. Professions that were identified included, a Nutritionist, Naturopath, Personal Trainer or Wellness Coach, with all authors reporting a single profession except for one blog where the author indicated multiple professional titles. The frequency of posting by authors varied (10–21 posts), however on average authors posted 15 times over the course of three months.

Successful healthy eating blogs were defined in this study as those listed on the first page of a major search engine (Google, Bing and Yahoo) [30‐32]. Healthy eating blogs identified by the search strategy did not include those authored by self-reported APDs. Despite healthy eating blogs being written by different authors, there were common characteristics between blog posts. These common characteristics included: (i) clearly conveying the purpose of each post, (ii) an understanding of the reader, (iii) consistent use of writing style, vocabulary and layout; and (iv) communicating healthy eating information in a practical manner (see Table 1).

Most blog posts focused on a single targeted healthy eating message, directly in line with the purpose of the post, rather than containing multiple healthy eating messages. Authors communicated the purpose early in the blog post using catchy headings and within the first few sentences of the first paragraph. Blog post headings were in the style of practical ways to increase fruit and vegetable intake; simple time saving food preparation recommendations; and practical ways to improve or add excitement into mealtimes. Readers were given a clear sense of the purpose of the post and this established expectations.

Findings indicated that authors conveyed a sense of understanding within blog posts to their reader’s needs. This was identified by the author directly acknowledging reader comments and concerns as the basis for a particular post. Communication was also encouraged by the author through statements that encouraged and directed the reader to provide feedback using the comment box. However, despite author encouragement, there were differences between reader participation using the comment box between blogs (see Table 2). Reader participation using the comment box ranged from 14 to 100%, with only Blog E achieving reader participation in all blog posts posted.

The writing style, vocabulary and layout of posts also appeared to encourage a relationship between the author and the reader. Posts were written in a conversational manner, with technical jargon explained using metaphors and simple language. Authors commonly wrote in the first person and positioned themselves as similar to their readers as an individual who also faces positive and challenging food and nutrition-related experiences. Example statements include: ‘… I’m passionate about healthy eating …’ (Blog A) and ‘I’ve personally watched [removed to de-identify]’ (Blog B). Both the conversational writing style and use of simple vocabulary complemented the translation of nutrition knowledge by allowing readers to understand nutrition-related concepts in non-technical language. In addition, blog posts focused on conveying a positive message about the benefits the food(s), rather than framing topics in negative terms.

Blogs incorporated the use of a consistent visual layout and structure from post to post within each blog. Commonly the visual layout of a post guided the reader’s attention from a bolded heading (emphasising the purpose of the post) to a complementary eye-catching photo directly below. The reader was guided towards the introduction of the post which was usually accompanied by another visual cue separating the introduction from the body of the post. The body of the post commonly focused on communicating healthy eating information. The reader was then guided to the end of the post which was identified by the authors’ sign off and encouragement to communicate via the comment box.

It was found that 64% of blog posts communicated healthy eating information in a practical manner by using recipes, practical tips and author recommendations (see Table 3). Typically, these techniques would be integrated into the structure of a post. Posts would commonly begin with an introductory paragraph which would state the purpose and/or associated health benefit(s). This would lead onto the next paragraph which would incorporate personal narratives or opinions from the author’s perspective, with practical information then provided. Recipes were used to communicate healthy eating messages in several ways including: a solution for readers to incorporate a specific food with health benefits into their diet; providing healthier alternatives to a dessert or snack; and food alternatives catering for a variety of food preferences. Recipes were influenced by season and the latest food trends, for example: coconut oil, maple syrup, rice malt syrup, avocado, organic coconut sugar, spelt flour and apple cider vinegar were incorporated into recipes. Vegetarian recipes were commonly posted along with current trends and ‘superfoods’, including ‘bowls’ (e.g. Buddha bowls, smoothie bowls), ‘smoothies’ (e.g. green smoothies, matcha smoothies) and ‘guilt free’ desserts (paleo bars, raw bars); further highlighting the up-to-date nature of blog posts. Practical tips and recommendations were commonly communicated in a straight-forward manner by bullet points or sequential steps. Recommendations were not exclusive to nutrition and often incorporated behavioural and motivational recommendations. For example, ‘Make healthy swaps’ (nutrition-related), ‘Inspire your friends and family’ (motivational) and ‘Don’t deprive yourself’ (behavioural) (Blog D).

A comparison between healthy eating information from blog posts with recommendations from the Australian Dietary Guidelines (ADG) suggested that only 43% of information clearly aligned with recommendations (see Additional file 1: Table S1). Most of the healthy eating information aligned with guidelines two and three, ‘Enjoy a wide variety of nutritious foods from the five food groups every day’ and ‘Limit intake of foods containing saturated fat, added salt, added sugars and alcohol’, respectively [42]. Posts often encouraged the consumption of fruit, vegetables, legumes and lean meats; and encouraged the concept of moderation, whilst permitting the occasional consumption of foods that do not fall within the five food groups.

Blogs are fast becoming a popular medium to access and communicate healthy eating information [1, 6, 16, 38, 54]. Although more research is needed to assess the effectiveness and impact of healthy eating blogs on behavioural change, it is recommended that the dietetic profession embrace this medium as an avenue to disseminate current, evidence-based and practical healthy eating information. To do so, the development of blog guidelines may be useful in providing dietitians – and potentially health professionals in other disciplines - with a framework to create long-term successful blog posts. Guidelines should consider findings from this study which suggest that successful healthy eating blogs should clearly convey the purpose of each post, understand reader interest by clarifying their needs, maintain consistency of writing style, vocabulary and layout, and communicate nutrition messages in a variety of practical and simple ways.