I still remember, very clearly, the first time the cost of insulin became real for me in Melbourne. I was at a pharmacy to fill a prescription for my rapid-acting insulin, and the pharmacist said, “That’ll be 180 dollars. Cash or card?” I paid, feeling numb, and by the time I stepped outside I was already reworking every expense in my budget around insulin. Shock first, then helplessness, then a quiet kind of fear.

I was diagnosed with type 1 diabetes when I was 16. My body doesn’t make insulin, so I inject it every day, and always will. This year I moved from India to Melbourne to do my Master’s. Australia’s diabetes care system is one of the best in the world, and as an international student, I’m outside it.

I knew that before I came. I bought my student health cover in India when I got my visa, and I knew it would never fully cover my insulin. I chose to come anyway. So this is not a complaint. Student visa holders aren’t covered by Medicare, Australia’s public health insurance scheme; it is a known condition of the visa, and this story is not an argument that it should change. It is a look at what that rule means in practice, for a condition you can’t pause.

The story moves from the world to one pharmacy counter: how long people live with type 1 in different places, how many people are missing, how many people like me live between two systems, and what the same insulin costs depending on which card is in your wallet.

Where you live decides how long you live

Type 1 diabetes is the same autoimmune condition everywhere. What changes is whether you can get insulin, glucose monitoring and a doctor who knows the condition, and the gap that makes is measured in decades. From what I have seen in India, government hospitals provide insulin free or at subsidised prices, but for most people it is still expensive. I was lucky: my family could afford my insulin and everything else my diabetes needed. Many families can’t, and that is what the number for India below reflects.

Figure 1. How to read this: each circle is a country. The further right, the longer people there live in general; the higher up, the more years a child diagnosed with type 1 loses. Hover for details. Type a country in the search box (or click a circle) to highlight it; “Clear” or double-click resets.
Takeaway: the years lost track the level of care a country can provide. A child diagnosed in India loses about 35 years, in line with other lower-middle-income countries; in Australia, about 10.

The people who never show up in the numbers

My own diagnosis was straightforward: my family took me to a doctor as soon as the symptoms started. For many people it isn’t. Type 1 is often missed, or mistaken for something else, and where care is hardest to reach, many die before anyone names the illness. The estimates call them “missing”. It feels especially unfair because of how insulin began: its discoverers sold the patent to the University of Toronto for $1, so that it could reach everyone who needed it. As Frederick Banting put it, “Insulin does not belong to me, it belongs to the world” (University of Toronto, n.d., para. 1).

Figure 2. How to read this: each bar is 100 people who should be living with type 1. The solid part is those who are; the pale part, with the number, is those who are missing because type 1 shortened their lives or was never diagnosed. Use the buttons to compare by income level or by world region; hover for exact numbers and uncertainty.
Takeaway: about 45 in every 100 are missing in India, close to the lower-middle-income average of 43. In Australia it is 13, below even the high-income average of 20. The difference is access to diagnosis and care.

Living between two systems

Australia is on the right side of those charts. My visa let me bring three months of medication from India; since then, I buy everything here. Someone I know who has permanent residency gets the same supplies at heavily subsidised prices. That is the line between the two systems, and a growing number of people live on my side of it, because they are here on a student visa. India is now one of the largest sources of international students in Australia.

Figure 3. How to read this: each line is one country’s international student enrolments per year; India is in orange. Use the menu to switch between all courses, university or vocational courses, or Victoria only. Hover for numbers and each country’s share of all international enrolments.
Takeaway: enrolments from India grew from about 27,000 in 2005 to about 181,000 in 2025, second only to China; Victoria has more Indian student enrolments than any other state. These students live under the same visa health rules I do. And because Australia’s type 1 diabetes figures come from the National Diabetes Services Scheme (NDSS) register, which needs Medicare, those of us who have type 1 are not counted in them (Ogle et al., 2025, supplementary Appendix 2).

The same insulin, four different prices

In Australia, medicines are subsidised through the Pharmaceutical Benefits Scheme (PBS), and diabetes supplies through the NDSS. Both depend on Medicare (National Diabetes Services Scheme, n.d.). Without it, you pay the pharmacy’s price, and private student health cover (OSHC) only refunds part of it, after a waiting period.

I use NovoRapid (insulin aspart) and Optisulin (insulin glargine). Without Medicare, my pharmacy charges me about $180 for one box of five cartridges of each. In India, I paid about ₹4,500 for the same box of NovoRapid, roughly A$67 at the late-October 2026 exchange rate. The chart below uses the PBS’s own full price, which covers five boxes, so it is the most conservative comparison possible: what I actually pay per box is several times higher.

Source: Department of Health, Disability and Ageing (2026), PBS Schedule, October 2026, via the PBS public API: dispensed price for maximum quantity (DPMQ) and patient co-payments. Without PBS access the DPMQ is used as the price; this is a lower bound, as pharmacies may charge more. OSHC rule: Bupa (2026), OSHC cover summary: refund of up to $50 per script above the $25 co-payment, up to $500 a year, after a 12-month waiting period for pre-existing conditions. Other OSHC policies differ.

Figure 4. How to read this: each row is one insulin product on the PBS; each shape is what one prescription costs a different person. Hover for the brand, how many units a prescription holds and the cost per 100 units; click a legend entry to hide or show a group. Use the buttons to switch between the price of a whole prescription and the average price per box, which also shows what I paid for one box.
Takeaway: a concession card holder pays $7.70 and a general patient $25 for every product. Without PBS access, even the PBS’s full dispensed price ($85 to $409) is only a minimum: my own pharmacy charges about $180 for one box of NovoRapid, a fifth of the quantity this PBS price covers. My OSHC refunds up to $50 of that, but only after a 12-month waiting period for pre-existing conditions, and only up to $500 a year. Even after the wait, that box would cost me about $130, still 26 times the $5 a box a general patient pays.

Over a year, the gap adds up

One prescription lasts weeks to months, depending on your dose. Here is what a whole year of insulin costs at PBS prices, the minimum anyone pays. I plan my spending around it: I rarely eat out, and I buy only what I need. On top of insulin, I spend about $40–50 a month on test strips and needles, almost as much as I spend on groceries.

Source: author’s calculation from the PBS Schedule, October 2026 (Department of Health, Disability and Ageing, 2026): units per prescription and DPMQ for each product; PBS co-payments. OSHC refund rule from Bupa (2026). Assumes half of daily insulin is long-acting and half rapid-acting, and ignores priming and wastage, so real costs are higher. Sensors, strips and needles are excluded because no official retail price list exists; NDSS registrants pay a co-payment of $17.35 per FreeStyle Libre 3 Plus sensor (National Diabetes Services Scheme, 2025).

Figure 5. How to read this: drag the slider (or use the arrow keys) to a daily dose; the dotted line and the list under the chart show what a year of insulin costs each person at that dose. Use the menu to switch between common combinations of insulin; hover to compare all four at any dose.
Takeaway: at 50 units a day with Optisulin and NovoRapid pens, a year of insulin costs a general patient about $100 and a concession card holder about $31. Without the PBS it is about $700 in my first year, and about $500 after the waiting period. These are PBS minimums; at the price I actually pay per box, a year would cost several times more. And that is before glucose sensors, test strips and pen needles, which the NDSS subsidises for Medicare holders but not for me.

For the next student with type 1

Australia is one of the best places in the world to live with type 1 diabetes, if you can get into its system. For a student visa holder, the rules are clear and published; what matters is planning around them. If you are coming to study here with type 1, these are the things worth planning for:

Nothing would have changed my mind about coming here. Australia is an amazing country, with lovely people and great opportunities, and I will always be grateful for the chance to study here. This isn’t a complaint. But in some corner of my mind, I hope that things become a little easier for students like me.

Data and methods

This story uses three independent data sources, each loaded in R from the original files:

  1. Global type 1 diabetes estimates (Figures 1–2): Ogle et al. (2025), Table 2 and supplementary Appendix 9, the International Diabetes Federation (IDF) Diabetes Atlas 11th edition and T1D Index version 3.0 estimates for 202 countries. Country estimates and their 95% confidence intervals were read from the Appendix 9 table in the supplementary Word file, and life expectancy without type 1 from Table 2 of the article; years of life lost = life expectancy without type 1 minus life expectancy with type 1. These are modelled estimates; data are sparse for many countries.
  2. PBS Schedule, October 2026 (Figures 4–5): Department of Health, Disability and Ageing (2026), downloaded from the PBS public data API (items, dispensing rules and co-payments for schedule 4751). Insulin items were selected by drug name and joined to their community-pharmacy dispensing rules to get each item’s DPMQ, pack size and maximum quantity.
  3. International student data, year to date June 2026 (Figure 3): Department of Education (2026). The records sit inside the workbook’s pivot-table cache, so they were read from it directly and summed by year, month, nationality, state and course sector. Full-year figures use the December year-to-date snapshot for 2005–2025.

Other sources are cited for rules and context, not plotted as data: the Bupa (2026) OSHC cover summary (refund amounts and waiting period), and two National Diabetes Services Scheme pages on registration eligibility and the sensor co-payment (National Diabetes Services Scheme, n.d., 2025). Figures 4–5 are the author’s calculations: without PBS access, the price of a prescription is taken as its dispensed price for maximum quantity (DPMQ), a lower bound; Figure 5 assumes half of daily insulin is long-acting and half rapid-acting and ignores priming and wastage. All visualisations were made in R (R Core Team, 2025) with plotly (Sievert, 2020) and htmlwidgets (Vaidyanathan et al., 2023), in an R Markdown document (Xie, 2015; Xie et al., 2018). Data were prepared with dplyr (Wickham, François, et al., 2026), tidyr (Wickham et al., 2025), stringr (Wickham, 2025b), stringi (Gagolewski, 2022), readr (Wickham, Hester, & Bryan, 2026), xml2 (Wickham, Hester, & Ooms, 2026), pdftools (Ooms, 2026) and httr2 (Wickham, 2025a).

Acknowledgements

PBS data: Copyright © 2024 Department of Health, Commonwealth of Australia. Permission is granted to use and redistribute this content, as long as all copyright statements are retained. Ogle et al. (2025) is published under a CC BY-NC-ND 4.0 licence.

Use of generative AI. I acknowledge the use of Claude (Opus 5.5 version), a large language model made by Anthropic (Anthropic, 2026; https://claude.ai), in producing this story. Between 01 and 27 October 2026, I used it to explain data visualisation concepts and RPubs publishing, and to check my writing for spelling and grammar. Prompts I used included: “please give me the steps to publish this in RPubs”; “could you check if there are any spelling or grammar issues in this paragraph?”. I then reviewed the outputs. The topic, the story, the personal experiences, and the visualisations are my own; I chose the data and the design direction, and I take responsibility for the final work.

References

Anthropic. (2026). Claude (Opus 5.5 version) [Large language model]. https://claude.ai

Bupa. (2026). Cover summary: Overseas student health cover [Brochure]. https://www.bupa.com.au/-/media/dotcom/files/oshc/migrated/oshc-cover/overseas_student_health_cover-product-info.pdf

Commonwealth Ombudsman. (n.d.). Overseas student health cover. PrivateHealth.gov.au. Retrieved October 26, 2026, from https://privatehealth.gov.au/health_insurance/overseas/overseas_student_health_cover.htm

Department of Education. (2026). International student data: Year-to-date (YTD) June 2026, all data [Data set]. Australian Government. https://www.education.gov.au/international-education-data-and-research/international-student-monthly-summary-and-data-tables

Department of Health, Disability and Ageing. (2026). Pharmaceutical Benefits Scheme (PBS) Schedule, October 2026 [Data set]. Australian Government. https://data.pbs.gov.au/api/pbs-api.html

Gagolewski, M. (2022). stringi: Fast and portable character string processing in R. Journal of Statistical Software, 103(2), 1–59. https://doi.org/10.18637/jss.v103.i02

National Diabetes Services Scheme. (n.d.). Registration. Retrieved October 23, 2026, from https://www.ndss.com.au/about-the-ndss/registration/

National Diabetes Services Scheme. (2025, November 17). FreeStyle Libre 3 Plus subsidised through the NDSS from December 2025. https://www.ndss.com.au/news/freestyle-libre-3-plus-subsidised-ndss-december-2025/

Ogle, G. D., Wang, F., Haynes, A., Gregory, G. A., King, T. W., Deng, K., Dabelea, D., James, S., Jenkins, A. J., Li, X., Ma, R. C. W., Maahs, D. M., Oram, R. A., Pihoker, C., Svensson, J., Zhou, Z., Magliano, D. J., & Maniam, J. (2025). Global type 1 diabetes prevalence, incidence, and mortality estimates 2025: Results from the International Diabetes Federation Atlas, 11th edition, and the T1D Index version 3.0. Diabetes Research and Clinical Practice, 225, Article 112277. https://doi.org/10.1016/j.diabres.2025.112277

Ooms, J. (2026). pdftools: Text extraction, rendering and converting of PDF documents (Version 3.9.1) [Computer software]. https://CRAN.R-project.org/package=pdftools

R Core Team. (2025). R: A language and environment for statistical computing (Version 4.5.2) [Computer software]. R Foundation for Statistical Computing. https://www.R-project.org/

Sievert, C. (2020). Interactive web-based data visualization with R, plotly, and shiny. Chapman and Hall/CRC. https://plotly-r.com

University of Toronto. (n.d.). Toronto’s gift to the world. Insulin 100. Retrieved October 25, 2026, from https://insulin100.utoronto.ca/?p=1360

Vaidyanathan, R., Xie, Y., Allaire, J. J., Cheng, J., Sievert, C., & Russell, K. (2023). htmlwidgets: HTML widgets for R (Version 1.6.4) [Computer software]. https://CRAN.R-project.org/package=htmlwidgets

Wickham, H. (2025a). httr2: Perform HTTP requests and process the responses (Version 1.2.2) [Computer software]. https://CRAN.R-project.org/package=httr2

Wickham, H. (2025b). stringr: Simple, consistent wrappers for common string operations (Version 1.6.0) [Computer software]. https://CRAN.R-project.org/package=stringr

Wickham, H., François, R., Henry, L., Müller, K., & Vaughan, D. (2026). dplyr: A grammar of data manipulation (Version 1.2.1) [Computer software]. https://CRAN.R-project.org/package=dplyr

Wickham, H., Hester, J., & Bryan, J. (2026). readr: Read rectangular text data (Version 2.2.0) [Computer software]. https://CRAN.R-project.org/package=readr

Wickham, H., Hester, J., & Ooms, J. (2026). xml2: Parse XML (Version 1.6.0) [Computer software]. https://CRAN.R-project.org/package=xml2

Wickham, H., Vaughan, D., & Girlich, M. (2025). tidyr: Tidy messy data (Version 1.3.2) [Computer software]. https://CRAN.R-project.org/package=tidyr

Xie, Y. (2015). Dynamic documents with R and knitr (2nd ed.). Chapman and Hall/CRC. https://yihui.org/knitr/

Xie, Y., Allaire, J. J., & Grolemund, G. (2018). R Markdown: The definitive guide. Chapman and Hall/CRC. https://bookdown.org/yihui/rmarkdown