The portrayal of diabetes in the Indonesian online press: a corpus-based discourse study

Abstract

Diabetes poses a substantial public health concern in Indonesia, standing as the third leading cause of mortality following stroke and heart disease. The media holds potential as a crucial tool for public education, influencing perceptions of health issues and dispelling stigmatization or misinformation surrounding individuals with diabetes. This article investigates the representation of diabetes in Indonesian online news articles by drawing on a corpus-assisted discourse studies framework. The study focuses on analyzing nominal and verbal collocates of diabetes. The findings reveal prevalent collocations, such as penderita (sufferer), pasien (patient) and penyakit (disease), which carry stigmatizing and derogatory connotations. Additionally, the study finds a predominant emphasis on behavioral discourses in discussions about diabetes, potentially contributing to societal stigmatization, while neglecting structural or societal representations. The media’s inclination toward individualized narratives may be driven by considerations of newsworthiness and a perceived need to enhance public awareness regarding behavioral actions linked to diabetes. The study emphasizes the importance of achieving a more balanced coverage of diabetes in Indonesian news media. Integrating portrayals that highlight structural issues into news narratives is crucial for fostering a comprehensive understanding, garnering public support for societal-level policy changes and reducing the stigma associated with people with diabetes.

Keywords:

• Diabetes
• news media
• discourse analysis
• corpus linguistics
• inclusive societies

REVIEWING EDITOR:

• Jeroen van de Weijer, College of International Studies, Shenzhen University, Shenzhen, China

SUBJECTS:

• Health Communication
• Cultural Studies
• Language & Linguistics

1. Introduction

Diabetes presents a significant public health challenge in Indonesia, ranking as the third most lethal ailment after stroke and heart disease (Nababan et al., 2023). With a reported prevalence affecting 19.5 million individuals in 2021 and projections indicating a surge to 28.6 million cases by 2045, Indonesia is anticipated to be among the top five nations with the highest prevalence of diabetes mellitus (Sun et al., 2022). Recent evaluations by Cho et al. (2018) emphasize the substantial global burden posed by diabetes prevalence, mortality rates and healthcare expenditures.

Recognized as influential entities with social, political and economic sway, news media platforms hold the power to spotlight and raise awareness on health-related issues (Finnegan & Viswanath, 1999; Jarlenski & Barry, 2013). Through their extensive reach, popularity and cost-effectiveness, mass media campaigns have become significant instruments in promoting health and disease prevention (Randolph & Viswanath, 2004). News media also has the potential to facilitate equal access to healthcare, disseminate accurate health information and knowledge, transform health attitudes and values, and encourage the adoption of new health behaviors (Venkat & Janakiram, 2021). In the context of diabetes, media can play a vital role in educating the public, shaping perceptions of illnesses and countering stigmatization or the spread of myths and misconceptions about people with diabetes (Browne et al., 2014; Douchinsky, 2017). As Bailey and McCrossin (2016) stated, the language employed by the media when discussing diabetes holds a significant impact on public views and attitudes toward the condition. This influence extends beyond people living with diabetes to encompass their spouses, friends, parents and other significant individuals within their social circles (Bailey & McCrossin, 2016), including policymakers (Bednarek, 2020).

A substantial amount of research has been conducted on how news media portrays diabetes. Hellyer and Haddock-Fraser (2011) examined the portrayal of type 2 diabetes in five leading British newspapers and their Sunday equivalents, while other studies drew on how news media construct diabetes in Australia (Bailey & McCrossin, 2016; Bednarek & Carr, 2020) and New Zealand (Gounder & Ameer, 2018). The majority of studies examining news coverage of diabetes typically use non-linguistic approaches, such as content and framing analysis (Gounder & Ameer, 2018; Hellyer & Haddock-Fraser, 2011). However, it is noteworthy that, as of present, no research has been conducted within a developing nation such as Indonesia despite the elevated prevalence of diabetes within its population. Most of the conducted studies have primarily emphasized aspects such as prevalence, risk factors, management and prevention (Soewondo et al., 2013; Tanoey & Becher, 2021; Widyahening et al., 2017). This creates a significant gap in our understanding of how diabetes is comprehended and presented by the news media in Indonesia.

Considering the urgency of the issue and its increasingly impactful consequences, it becomes essential to engage in more in-depth research regarding the Indonesian news media representation of diabetes. This study, therefore, aims to fill this gap in the literature by investigating the representation of diabetes in Indonesian online news reports, employing a corpus-assisted discourse studies (CADS) framework (Baker et al., 2008; Partington et al., 2013). The resultant findings have the potential to inform and enhance the approaches employed in the coverage of diabetes-related matters by Indonesian news media, thereby fostering improvements in their practices.

2. Literature review

2.1. Corpus-assisted discourse analysis

The present study adopts the synergy of corpus linguistic techniques and discourse analysis to identify how online media in Indonesia represents diabetes (Baker, 2023). The approach can quickly and accurately carry out complex calculations on the linguistic data (Subtirelu & Baker, 2017). By using a corpus linguistic approach, researchers are able to analyze large amounts of data. As stated by Baker (2006), this method may be able to minimize writer bias and offer a holistic analysis of texts. By focusing on every word, phrase and sentence structure, this approach can provide a profound understanding of the meanings embedded in the discourse. The use of a linguistic corpus allows researchers to identify common sentence patterns in the discourse about diabetes. As explained by Biber et al. (2006), CADS is grounded in empirical evidence and involves analyzing language patterns found in natural texts. Moreover, this method can provide insights into changes and perspectives that may emerge in society regarding diabetes. In this way, CADS analysis can help uncover the dynamics and development of the diabetes issue in the realm of media, enriching our understanding of how this disease is represented and discussed in society.

Building upon previous studies that utilized CADS approaches (e.g. Baker et al., 2008; Fajri, 2017), the synergy of corpus linguistics and discourse analysis has proven effective in unveiling how the media depicts specific topics, including health. For instance, Brookes and Baker’s (2022) study extensively explores how obesity and its associated risks are portrayed in the print media, including both tabloids and broadsheet newspapers, in the United Kingdom. The research underscores that obesity is recognized as a risk factor for various diseases, including cancer. However, the study also reveals that obesity is perceived as an outcome of certain risk factors, such as overeating or insufficient sleep, indicating the intricate nature of the discourse surrounding obesity and the interplay between risk and consequences. Similarly, Fajri et al. (2023) investigated how the media in Indonesia portrays obesity. The findings indicate that obesity in the media context is predominantly perceived as a personal responsibility. In other words, each person is seen as entirely responsible for minimizing the risk of obesity and addressing health issues associated with weight management. This perspective promotes a narrative that places control and direct action on individuals, emphasizing the role of personal responsibility in managing health issues while neglecting the structural and environmental influences contributing to this health problem. These studies demonstrate the potential of CADS in dissecting media discourse and shedding light on societal perceptions and attitudes toward health issues like obesity.

2.2. News media representations of diabetes

Several studies have explored the portrayal of diabetes in news media. Hellyer and Haddock-Fraser (2011) conducted content analysis on three months of media coverage of diabetes in the UK, revealing a portrayal of diabetes as a behavioral issue, with obesity being highlighted as a contributing factor. Foley et al. (2020) observed a shift in framing over time, from medical to behavioral and eventually to societal, within the context of the obesity epidemic. Their findings underscored how framing diabetes influences public perceptions, biomedical approaches and broader societal implications. Similarly, research on North American news by Gollust and Lantz (2009) found that obesity and behavior were central themes in the representation of diabetes. Stefanik-Sidener (2013) analyzed media framing of diabetes in the New York Times and identified a predominant medical or behavioral frame, with limited use of a societal perspective across different types of diabetes. These findings resonate with those of Gounder and Ameer (2018), which reveal that the New Zealand print media constructs diabetes as an individual’s medical concern with frequent references to obesity and behavioral choices as causal factors. In the Australian context, Bailey and McCrossin (2016) investigated the language used by the Australian print news media to depict diabetes. Their study revealed that there was no significant increase in articles solely using preferred language or neutral language, suggesting that most news texts canonically utilized both preferred and avoided language.

CADS has also been used to examine news media representations of diabetes. Bednarek and Carr (2020) investigated how diabetes is reported in Australian newspapers, concentrating on the volume of articles, the mismatch between such coverage and the actual societal prevalence of diabetes, and the particular use of language. The findings reveal that the portrayal of diabetes in Australian newspapers does not entirely correspond with the true occurrence of diabetes among the population. The study highlights issues in language use, such as referring to individuals with diabetes as ‘diabetics’, and underscores the insufficient representation of Aboriginal and Torres Strait Islander communities in diabetes-related reporting.

The current research distinguishes itself from preceding studies by scrutinizing discourses pertaining to ‘diabetes’ within news texts expressed in a language other than English, specifically within the context of a developing country, Indonesia. This unique focus introduces distinct contextual elements to the existing body of research.

3. Methodology

3.1. Data source and data collection

The study utilized a dataset consisting of online news articles in Indonesia that discussed diabetes. These articles were gathered from two prominent online media platforms, Detik.com and Kompas.com, throughout the year 2023. The decision to focus on these platforms was driven by their recognition as the top three most visited Indonesian news websites (Similarweb, 2023). This intentional selection aimed to ensure the representation of diverse perspectives and comprehensive coverage of information regarding diabetes in the Indonesian online media landscape.

The collection of articles involved searching for the term ‘diabetes’ in Detik.com and Kompas.com. Articles containing at least one mention of diabetes were included in this study. However, we imposed a limitation on the publication year of the articles, ensuring that only articles published within the desired timeframe were included in the study. Subsequently, from all the collected articles, we randomly selected 100 articles from Detik.com and 100 articles from Kompas.com. This approach ensures diversity and representativeness in the selection of articles to be analyzed while still adhering to specific criteria related to diabetes and the publication year. Thus, these steps lead to a more focused and representative data collection for the research purpose regarding the coverage of diabetes in the online media of Detik.com and Kompas.com.

3.2. Data analysis

For the analysis, we utilized a combination of quantitative and qualitative methods (Creswell, 2014), integrating both approaches to analyze the depiction of the term ‘diabetes’ in the online media. The quantitative involves employing linguistic corpus techniques to examine statistics and the frequency of the word ‘diabetes’ in the context of discourse. On the other hand, the qualitative approach was applied when interpreting the findings of the corpus analysis, examining collocations and concordances. We used AntConc 4.2.4 software (Anthony, 2023) to process the data. The result indicates that the corpus data of news articles has a total word of 89,727. The software was also used to conduct collocation and concordance analyses.

Qualitative analysis in this study involves two main stages. Firstly, an exploration of the collocation patterns of the term ‘diabetes’ in the corpus was conducted. Collocation, as defined by Stubbs (2001), refers to the repeated and meaningful co-occurrence of words. Baker (2023) highlights the advantages of employing collocational analysis in discourse analysis, as it can reveal prominent and distinctive lexical patterns related to a specific subject. These patterns serve as valuable resources for gaining insights into discourses associated with that subject. Subsequently, we analyzed concordances based on collocated words categorized as content words, as they provide a more specific or specialized meaning in a particular context (Pennebaker et al., 2014). According to Baker (2010), a concordance is essentially a compilation of all instances of a word, phrase or search term, along with contextual information surrounding each occurrence. Analyzing concordances is crucial for discourse analysis for two main reasons: firstly, they help identify evidence of different ‘prosodies’ or ‘preferences’, and secondly, the absence of concordance analysis may lead to incorrect assumptions about the corpus content. Concordance analysis in this study focuses on noun and verbal collocates, as well as the semantic prosody of those words. Semantic prosody, as termed by Stubbs (2001), is identified as discourse prosody, a characteristic extending beyond a single unit in a linear sequence. Baker (2010) adds that prosodies can be uncovered through the examination of a list of strong collocates associated with a word. In this research, a span of three words on each side of the node word is employed. Log-likelihood (Dunning, 1993) is chosen as the statistical method, as it evaluates collocations based on significance, assigning a higher score to a non-random association and avoiding high scores for words with relatively low frequencies.

4. Findings

The initial phase of this study involves identifying the collocation patterns related to diabetes. The emphasis is placed on nominal and verbal collocations, as they prove valuable in discourse analysis. Analyzing these collocations allows researchers to evaluate the portrayal of diabetes in the text corpora of Detik.com and Kompas.com. Building upon the criteria outlined in the preceding section for collocation analysis, 20 nominal collocates and 10 verbal collocates were generated. Nevertheless, this study only focused on the top 10 nominal and verbal collocates.

4.1. Nominal collocates of diabetes

Following the analysis of corpus data in the AntConc, it was observed that diabetes most frequently occurs in association with nouns. Therefore, the first step taken involves analyzing nominal collocations with the highest log-likelihood scores (Table 1).

Table 1. Nominal collocates of diabetes.

No.	Words	Observed collocate frequency	Log-likelihood
1.	Penderita (sufferer)	465	1072.827
2.	Tipe (type)	377	900.490
3.	Risiko (risk)	177	206.025
4.	Pengidap (sufferer)	94	204.534
5.	Penyakit (disease)	207	184.683
6.	Melitus (mellitus)	61	144.421
7.	Pasien (patient)	80	116.276
8.	Kasus (case)	61	91.606
9.	Gula (blood sugar)	98	80.263
10.	Gejala (symptom)	78	68.252

Three of the nominal collocates of diabetes indicate words used to refer to individuals with diabetes, which are penderita (sufferer), pengidap (sufferer) and pasien (patient). Specifically, the term penderita (sufferer) is the most prevalent, occurring 465 times, followed by pengidap (sufferer), which is observed 94 times, and pasien (patient), with a frequency of 80 occurrences. These terms are commonly employed to convey the notion that an individual is suffering from a specific condition, typically associated with an ailment or an adverse circumstance (see Extracts 1–3). Given that all three terms, penderita (sufferer), pengidap (sufferer) and pasien (patient), are associated with adverse experiences, these collocations contribute to negative discourse prosody surrounding the word diabetes. These words have been identified as expressions to be avoided when discussing diabetes (Speight et al., 2012, 2021) because they are perceived by a considerable number of individuals with diabetes as labels that carry negative, discriminatory, stigmatizing and derogatory connotations (Dunning et al., 2017).

• (1) Selain itu, penderita diabetes tipe 2 yang kurang tidur mungkin menderita sensitivitas insulin yang lebih rendah. [Moreover, sufferers of type 2 diabetes who lack sleep may experience lower insulin sensitivity.] (Kompas.com, 14 October 2023)

• (2) Menurut National Health Service (NHS), seorang pengidap diabetes tipe 2 bisa hidup dengan penyakit tersebut selama bertahun-tahun tanpa menyadari. [According to the National Health Service (NHS), sufferers of type 2 diabetes can live with the condition for years without being aware of it.] (Detik.com, 15 November 2023)

• (3) Pasien diabetes sendiri memang memiliki gejala sering terbangun di sela-sela tidur karena keinginan untuk kencing yang sangat sering. [Diabetic patients themselves often experience symptoms of frequent awakening during sleep due to the need to urinate frequently.] (Kompas.com, 10 October 2023)

Naming individuals with diabetes as sufferers positions them as helpless victims, diminishing their agency to live well with diabetes (Speight et al., 2012). Similarly, the use of the word patient relegates the person to a passive role, implying a state of tolerance toward their condition, as they await the intervention of an expert or health provider (Speight et al., 2021), disempowering individuals to actively engage in managing their health. Rather than employing the aforementioned terms, the preferred designation is orang dengan diabetes (people with diabetes) (Bednarek & Carr, 2020; Speight et al., 2021). This phenomenon indicates that the media plays a crucial role in shaping public perceptions regarding individuals with diabetes (Douchinsky, 2017). The media functions not solely as a disseminator of information but also as a formative influence on opinions and public perspectives regarding specific health conditions, including diabetes.

Another collocation to be avoided is penyakit (disease), which co-occurs with diabetes 207 times (see Extract 4). The term ‘disease’ carries negative connotations, suggesting something potentially contagious, and individuals with diseases are often subject to avoidance or fear (Speight et al., 2012). It is crucial to recognize that diabetes is a chronic condition requiring lifelong management, rather than a disease to be treated (Speight et al., 2021). This distinction underscores the importance of framing diabetes as a condition that can be effectively managed rather than perpetuating misconceptions associated with the term disease.

• (4) Mereka yang memiliki riwayat penyakit diabetes dalam keluarga dianjurkan untuk menjalani pemeriksaan atau tes darah secara rutin. [Those with a family history of diabetes disease are advised to undergo regular blood tests or screenings.] (Detik.com, n.d. 2023)

The term risiko also frequently co-occurs with diabetes, appearing 177 times. Defined as an unpleasant consequence (Aven, 2012), its contextual usage in relation to diabetes indicates factors associated with elements that either elevate or mitigate diabetes symptoms (see Extracts 5 and 6). By framing the risk of diabetes, the media assumes a pivotal role in shaping public perceptions concerning prevention strategies for diabetes (Stefanik-Sidener, 2013), enhancing public understanding of health maintenance and reducing the likelihood of disease contraction. However, it should be noted that this approach may reinforce the medical or behavioral frames of diabetes.

• (5) Penelitian menunjukkan bahwa meminum kopi secara teratur memiliki kaitan dengan risiko diabetes tipe 2 yang lebih rendah. [The research indicates that regularly drinking coffee is associated with a lower risk of type 2 diabetes.] (Detik.com, 14 November 2023)

• (6) Dilansir dari laman Healthline, beberapa penelitian lama menemukan bahwa makan makanan yang digoreng meningkatkan risiko terkena diabetes tipe 2. [According to Healthline, several old studies suggest that consuming fried foods increases the risk of developing type 2 diabetes.] (Kompas.com, 9 October 2023)

The next collocate is kasus (case). The association of the term diabetes with kasus (case) not only signifies a worrying increase in diabetes cases in Indonesia but also emphasizes how the media focuses on this health issue. The concordance analysis shows that diabetes is commonly paired with kasus (case) to highlight the growing prevalence of this disease among the Indonesian population. This pattern observed in media reporting signifies an elevated awareness and concern regarding the diabetes epidemic in the country. Additionally, the frequent collocation of diabetes with kasus in news reports often involves a comparison of the prevalence of type 1 and type 2 diabetes. This comparative analysis holds significance as it contributes to an understanding of the distribution of different diabetes types within the population. Concordance analysis indicates that type 1 diabetes, although still present, is notably less common, accounting for only 5–10% of all diabetes cases in Indonesia (see Extract 7). Conversely, type 2 diabetes is more prevalent, suggesting that lifestyle factors and obesity, significant contributors to type 2 diabetes, are a major concern in the Indonesian context.

• (7) Diabetes tipe 1 juga jarang terjadi, hanya menyumbang lima hingga 10 persen dari seluruh kasus diabetes. [Type 1 diabetes is uncommon, constituting only five to ten percent of the total diabetes cases.] (Kompas.com, 18 September 2023)

Moreover, the utilization of the term kasus (case) in conjunction with diabetes in Indonesian media, particularly when illustrating the escalating trend of diabetes among children, signifies a troubling shift in the demographic affected by this chronic condition. Traditionally perceived as an ailment afflicting older adults, the growing incidence of diabetes in younger populations, including children, indicates changing dietary habits and reduced physical activity of children (see Extract 8).

• (8) Naiknya diabetes pada anak salah satunya dipicu metabolic syndrome seperti kelebihan berat badan yang disebabkan anak suka mengonsumsi makanan manis berlebih, intensitas olahraga yang rendah, dan perilaku sedentari yang tidak dibatasi. [The increase in diabetes in children is partly triggered by metabolic syndrome, such as excess weight caused by a child’s tendency to consume excessive sweet foods, low physical activity levels and unrestricted sedentary behavior.] (Kompas.com, 15 November 2023)

Diabetes also collocates with gula (mostly referring to blood sugar) and gejala (symptoms). The co-occurrence with gula (blood sugar) frequently features in articles, suggesting that maintaining optimal glucose levels is crucial in managing and monitoring diabetes. Regular monitoring of blood sugar levels helps individuals with diabetes make informed decisions about their diet, medication and lifestyle choices to keep their blood sugar within a target range. Meanwhile, the co-occurrence with symptoms indicates that diabetes can manifest in various ways, such as increased thirst, frequent urination, unexplained weight loss and fatigue (see Extract 10).

• (9) Namun saraf tersebut dapat rusak, ketika kadar gula darah penderita diabetes tidak terkontrol dengan baik dan terus-menerus tinggi. [However, these nerves can become damaged when the blood sugar levels of diabetes patients are poorly controlled and consistently high.] (Kompas.com, 15 November 2023)

• (10) Diabetes dapat menyebabkan gejala seperti mudah haus, sering buang air kecil, berat badan menurun, lemas, dan lain sebagainya. [Diabetes can cause symptoms such as excessive thirst, frequent urination, weight loss, fatigue, and so on.] (Detik.com, 13 November 2023)

These insights underscore the substantial and intricate nature of the diabetes issue, emphasizing the necessity for targeted preventive measures, a comprehensive understanding of the various types of diabetes and the influence of lifestyle in mitigating diabetes risks. This aligns with the perspective proposed by Stefanik-Sidener (2013) that the media not only serves to raise awareness about diabetes but also plays a role in disseminating information regarding prevention, management and treatment. However, akin to risiko (risk), the collocates kasus (case), gula (blood sugar) and gejala (symptom) also contribute to the prevalence of medical and behavioral frames of diabetes, consistent with the findings of Gollust and Lantz (2009) and Stefanik-Sidener (2013). This framing tends to concentrate on the medical and behavioral or lifestyle factors associated with the causes and solutions of diabetes, often neglecting more intricate policy or societal initiatives (Stefanik-Sidener, 2013).

4.2. Verbal collocates of diabetes

The word diabetes significantly collocates with 10 verbs, as illustrated in Table 2. It predominantly collocates with the verbs menderita (suffer) and mengidap (suffer). Similar to their corresponding noun forms, these verbs carry comparable meanings and can be employed to signify that an individual is undergoing a specific condition, whether it pertains to the physical, emotional or situational domain.

Table 2. Verbal collocates of diabetes.

No.	Words	Frequency	Log-likelihood
1.	Menderita (suffer)	53	100.454
2.	Mengidap (suffer)	51	93.171
3.	Terkena (has/get/develop)	55	92.730
4.	Mencegah (prevent)	61	54.891
5.	Adalah (is)	113	36.917
6.	Perlu (need)	54	33.090
7.	Didiagnosis (diagnosed)	11	21.557
8.	Berisiko (risk)	17	20.886
9.	Cegah (prevent)	9	20.416
10.	Mengelola (manage)	18	20.340

Concordance analyses of these verbs reveal that both menderita and mengidap are frequently utilized to portray challenging experiences, encompassing instances of both physical and emotional distress (see Extract 11). These portrayals also encompass descriptions of the cause and effect of diabetes, predominantly framed from a medical perspective (see Extract 12).

• (11) Kondisi sosok Panji Petualang sempat menjadi sorotan publik beberapa waktu lalu. Pasalnya ia tampak amat kurus lantaran sudah mengidap diabetes selama lima bulan. [The condition of the adventurer figure Panji has recently garnered public attention. This is because he appeared extremely thin due to having suffered from diabetes for five months.] (Detik.com, 20 September 2023)

• (12) Penderita diabetes seringkali menderita gangguan tidur dan minum kopi dapat memperburuk masalah yang mereka alami. [Diabetes sufferers often suffer from sleep disturbances, and consuming coffee can exacerbate the issues they face.] (Kompas.com, 14 October 2023)

Diabetes also collocates with the word terkena (get) which suggests that the condition of diabetes is something that someone ‘gets’ or ‘acquires’, implying a sense of being affected or afflicted by it. This verb choice emphasizes the passive nature of the condition, highlighting that it is something that happens to individuals rather than being actively chosen or pursued. Furthermore, the term diabetes is frequently linked with the verb mencegah (prevent) with a frequency of 61 occurrences, cegah (prevent), which appears 9 times, and mengelola (manage) with 18 instances. Since cegah is the base form of the word mencegah, they are considered as a single entity for analysis. Concordance lists of mencegah (prevent) and cegah (prevent) indicate the emphasis on preventive measures within the context of diabetes, while mengelola (manage) highlights the recognition that, for individuals already diagnosed with diabetes, effective management is essential for controlling blood sugar levels and mitigating potential complications (see Extracts 13 and 14).

• (13) Penelitian terbaru mengungkap, jalan kaki setelah makan bermanfaat menurunkan kadar gula darah dan mencegah munculnya diabetes. [A recent study has revealed that walking after meals is beneficial for lowering blood sugar levels and preventing the onset of diabetes.] (Kompas.com, 31 October 2023)

• (14) Kegiatan tersebut pun bisa memotivasi diabetesi agar kualitas hidupnya tetap terjaga dan mampu mengelola diabetes dengan baik. [Such activity can also motivate diabetics to maintain their quality of life and effectively manage their diabetes.] (Kompas.com, 15 August 2023)

The examination of the verbal collocate perlu (need) indicates a crucial aspect of discourse on diabetes management and care. It suggests a concentration on the requirements and necessities for individuals diagnosed with this condition. In this context, the concordance analysis unveils specific needs and actions that are essential for people living with diabetes, encompassing aspects such as insulin therapy, lifestyle changes, medication and hydration (see Extracts 15 and 16).

• (15) Penderita diabetes perlu melakukan terapi insulin, termasuk dengan menyuntikannya, dengan anjuran dokter agar kondisi yang dialami membaik dan dapat menghindari komplikasi. [People with diabetes need to undergo insulin therapy, including administering injections, as advised by a doctor in order for their condition to improve and to avoid complications.] (Kompas.com, 27 August 2023)

• (16) Namun, penderita diabetes tipe 2 perlu melakukan perubahan kebiasaan dan mengonsumsi obat-obatan tertentu sebelumnya. [However, people with type 2 diabetes need to make lifestyle changes and consume certain medications.] (Kompas.com, 27 August 2023)

• (17) Arya menyebut MWA menderita diabetes dan didiagnosis mengalami komplikasi asam urat dan infeksi. [Arya mentioned that MWA suffers from diabetes and has been diagnosed with complications including gout and infection.] (Detik.com, 18 July 2023)

• (18) Orang dengan asam urat tinggi berisiko mengalami diabetes begitupun sebaliknya. [Individuals with high uric acid levels risk developing diabetes, and vice versa.] (Kompas.com, 7 November 2023)

Diabetes also collocates with the words didiagnosis (diagnosed) and berisiko (risky). The verb didiagnosis suggests that individuals are being formally diagnosed with diabetes, indicating a medical confirmation of the condition (see Extract 17). Additionally, berisiko suggests that individuals are at risk of developing diabetes, highlighting the potential susceptibility to the condition based on various factors such as lifestyle, genetics or other health conditions (see Extract 18).

5. Discussion

The findings of the analyses shed light on the discourses surrounding diabetes in the Indonesian online news media. Several key points emerge from the analysis, which have implications for how diabetes is portrayed. Firstly, the prominence of nominal collocates such as penderita (sufferer), pengidap (sufferer) and pasien (patient) underscores the negative discourse prosody surrounding diabetes. These terms, often perceived as stigmatizing and disempowering, position individuals with diabetes as passive victims rather than active agents in managing their health. This aligns with previous research (e.g. Bednarek & Carr, 2020; Speight et al., 2021) advocating for more empowering and person-centered language, such as orang dengan diabetes (people with diabetes), to foster a more positive and inclusive discourse.

Similarly, the frequent collocation of penyakit (disease) with diabetes perpetuates misconceptions and negative associations, framing diabetes solely as a pathological condition rather than a chronic condition requiring lifelong management. Emphasizing the distinction between ‘disease’ and ‘condition’ is crucial in destigmatizing diabetes and promoting a holistic understanding of its management and implications (Speight et al., 2021).

Thirdly, the association of risiko (risk) and berisiko (risky) with diabetes underscores the importance of preventive measures and public health interventions, while the collocations involving gula (blood sugar) and gejala (symptoms) highlight the importance of glucose monitoring and symptom recognition in diabetes management. By depicting diabetes in terms of risk factors and prevention strategies, the media can contribute to raising awareness and promoting healthy behaviors. Through consistent exposure to specific terms related to diabetes prevention, the public may internalize this information, fostering an increased awareness of the measures they can take to maintain their health and prevent diabetes.

Nevertheless, it should be noted that the causation and treatment of diabetes, especially type 2 diabetes, are predominantly associated with dietary and lifestyle modifications, aligning closely with the behavioral discourses of diabetes. This may correspond to neoliberal discourses of personal responsibility, signifying that individuals bear full responsibility for sustaining acceptable blood sugar levels through adherence to a regimen of regular professional supervision, self-monitoring, dietary control, physical exercise and often medication (Broom & Whittaker, 2004; Gounder & Ameer, 2018; Hunt & Koteyko, 2015). Individuals are depicted as morally deviant in their health behaviors, attributing the cause of their health conditions to personal choices (Lawrence, 2004). Consequently, individuals with diabetes are susceptible to stigmatization from friends, other people with diabetes and health professionals (Broom & Whittaker, 2004).

In addition, the analyses of nominal and verbal collocates in this current study reveal a deficiency in the discussion of societal causes or policy solutions, indicating a lack of emphasis on the structural discourse (Gounder & Ameer, 2018) or societal framing (Stefanik-Sidener, 2013). This is in line with neoliberal discourses, which suggest that the government’s involvement in individuals’ health should be kept to a minimum (Horton, 2007). This encompasses the neglect of topics such as the availability of nutrition facts, access to safe parks, availability of fresh, healthy foods and access to medical care and diabetes education. Considering that type 2 diabetes impacts minority and low-income populations, discussions on these societal-level solutions should be particularly crucial (Stefanik-Sidener, 2013). In this way, the news media seem to fall short in fully capturing the complexity of the diabetes condition. A plausible explanation for the focus on cause-and-effect narratives utilizing individualized or behavioral frames is that the media aims to address what they perceive as a necessity to enhance public awareness regarding behavioral actions that contribute to diabetes, particularly type 2 diabetes (Gounder & Ameer, 2018). Another contributing factor could be newsworthiness, as news related to food, eating practices, physical activity and diets that can influence diabetes levels or risks is considered more newsworthy, rather than news focusing on structural issues (Fajri et al., 2023).

In light of this, the Indonesian news media may need to achieve greater balance in their coverage of diabetes, integrating more structural frames into their narratives. The inclusion of the structural frame can garner greater public support for policy changes at the societal level aimed at addressing the issue (Niederdeppe et al., 2013), fostering better public understanding of diabetes and reducing stigma for those who have it (Stefanik-Sidener, 2013).

6. Conclusion

This article highlights the significance of investigating how diabetes is portrayed in Indonesian online news reports. It utilizes a corpus-assisted discourse studies framework to analyze nominal and verbal collocates of the term ‘diabetes’. The findings indicate a concerning trend in the use of terms like penderita (sufferer), pengidap (sufferer) and pasien (patient) to refer to individuals with diabetes. These terms contribute to a negative discourse surrounding diabetes, implying a sense of helplessness and passivity among individuals with the condition. Importantly, such labels carry negative, stigmatizing connotations and diminish the agency of individuals to actively manage their health.

Additionally, the findings also demonstrate the media’s emphasis on behavioral discourses of diabetes, centered on personal choices and lifestyle modifications, raising concerns about the comprehensiveness of the narrative surrounding diabetes in Indonesia. To foster a more nuanced understanding of this complex health issue, the media should broaden their coverage to include societal-level solutions. Integrating structural representations into diabetes discourse not only enhances public understanding but also promotes support for policy changes that can address the root causes of the condition and reduce the stigma of people with diabetes.

Despite the valuable insights gained from this study, it is important to acknowledge its limitations. The corpus for this research was constructed from 100 articles, each of which was published by Detik.com and Kompas.com in 2023. While this sample provides a snapshot of media discourse during that specific timeframe, the findings may not capture potential variations over time or reflect evolving patterns in media representation. To address this limitation, future research endeavors could benefit from expanding the dataset by including a larger number of articles from different news portals and extending the period of article release years. A diachronic analysis, which examines changes in language and discourse over time, is recommended to uncover any shifts or trends in how diabetes is portrayed in Indonesian media.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Restu Santyarini

Restu Santyarini is a master’s student in Linguistics at the Faculty of Humanities, Universitas Airlangga, Indonesia. Her research interests include corpus linguistics, discourse analysis, and English language teaching.

Muchamad Sholakhuddin Al Fajri

Muchamad Sholakhuddin Al Fajri is a Lecturer in Linguistics at the Faculty of Vocational Studies, Universitas Airlangga, Indonesia. His research interests include corpus-assisted discourse studies, corpus linguistics, critical discourse studies, and English for academic purposes.