All strengths of ramipril (Tryzan), an ACE inhibitor, will run out over the coming months due to manufacturing issues. The 1.25 mg, 5 mg and 10 mg capsules are expected to be out of stock sometime in May and 2.5 mg capsules in July.

Alternative brands of 1.25 mg (Ramipril Viatris), 5 mg and 10 mg (Rampiril Mylan generics) strengths have been listed on the Pharmaceutical Schedule since 1^st May, however, neither brand is approved by Medsafe, therefore will need to prescribed for supply under Section 29A of the Medicines Act. An alternative brand of the 2.5 mg strength is not currently available.

Reassure patients that there has been no change to the active ingredient, but the medicine will be in tablet form (rather than capsule) and come in a smaller pack size: 28 (Mylan generics) or 30 (Viatris) tablets instead of 90 (Tryzan). The packaging of 5 mg and 10 mg strengths will also be in Italian with an English sticker.

Patient information sheets about the brand changes are available here

Supply of montelukast 5 mg chewable tablets (Montelukast Viatris), indicated for asthma prophylaxis and allergic rhinitis, may run out due to issues at the factory where it is made. A re-supply date is unknown; other strengths are not currently affected.

An alternative brand, Relonchem (S29), has been listed on the Pharmaceutical Schedule since 1^st May, however, it is not approved by Medsafe, therefore will need to be prescribed for supply under Section 29A of the Medicines Act. Reassure patients that there has been no change to the active ingredient and the tablet remains chewable, but the appearance will differ (pink in colour with a “5” etched on one side).

Olanzapine 5 mg and 10 mg orodispersible tablets (Zypine ODT), used in the treatment of various psychoses, are out of stock. Supply issues affecting 10 mg orodispersible tablets have been ongoing since February due to manufacturing and shipping delays (as reported in Bulletin 142). There is low stock of the alternative brands of the 10 mg strength that were listed in March (Olanzapina Mylan Pharma and Olanzapina Mylan).

Since 1^st May, two new brands have been listed on the Pharmaceutical Schedule: Apo-Olanzapine and Olanzapine ODT Viatris. Apo-Olanzapine 5 mg and 10 mg tablets are available to order; this brand is not approved by Medsafe, therefore will need to be prescribed for supply under Section 29A of the Medicines Act. Stock of Olanzapine ODT Viatris is not currently available.

There is a supply issue affecting stock of colestyramine 4 g powder sachets (Section 29, unapproved medicine), due to problems at the factory; it is expected that access issues may begin for patients from mid-May. A re-supply date is unknown; an alternative product is being sourced, but the details are yet to be confirmed. Colestyramine is a bile acid sequestrant used as a lipid-lowering treatment and for pruritus associated with partial biliary obstruction and primary biliary cirrhosis, as well as diarrhoea following ileal resection or ileal disease.

Updated cervical screening guidelines have been developed to replace the 2023 version; however, they have not yet been published in full. The updates have a range of implications, including for digital infrastructure, NCSP Register services, communications and provider training, which need to be addressed prior to full implementation. Some components of the updated guidelines have been released in advance, announced in a recent addendum. Most of these updates do not have major implications for clinical practice.

One change for primary care is that test of cure (i.e. HPV and cytology co-test) follow-up of people aged under 50 years with positive margins after excision treatment of high-grade squamous intraepithelial lesions (HSIL), and after excision treatment for adenocarcinoma in situ, can now occur in primary care rather than in a colposcopy clinic. Screening recommendations after gynaecological cancer are also covered, as well as a revised list of what constitutes immune deficiency. View all changes here.

The latest data show that overall cervical screening coverage is at approximately 76%; slightly below the Programme’s target of 80%. This is a timely reminder to opportunistically check whether eligible patients are up to date with cervical screening.

A bpac^nz clinical audit is available for identifying patients who are not participating in regular cervical screening.

A closer look at the data:

The National Cervical Screening Programme reported that of the almost 900,000 HPV primary screens that have been conducted since HPV testing was implemented:

• Approximately 82% were self-tests
• 2% were positive for HPV 16/18
• 8% were positive for HPV Other
• Approximately 16% were performed in people who were un- or under-screened with the previous cytology-based programme

There has been an increase in participation among all ethnic groups. For example, cervical screening coverage increased in Māori by 12.3% between August, 2023 (57.2%), and February, 2026 (69.5%). Pacific peoples had the largest increase in cervical screening participation (20.4% increase in this period).

Cervical screening coverage data are also available by district/region; check how your area is doing here.

Nasal membranes, as with the rest of the upper respiratory tract, are lined with secretory cells that produce mucus (e.g. goblet cells), and ciliated cells (epithelial cells with hair-like projections).¹ The mucus lining of the nasal cavity traps inhaled particles, viruses and bacteria. The ciliated cells move mucus down the nasal cavity to the nasopharynx where it is swallowed and enters the gastrointestinal tract (mucociliary clearance).^{1, 2} Up to one litre of mucus is produced each day, however, this volume can double in response to sinus or respiratory tract inflammation, e.g. during a respiratory infection.³ Thankfully, most of this is swallowed spontaneously.

During infection, mucus production is increased to improve trapping and clearance of pathogens, while mucociliary clearance is disrupted (e.g. due to pathogen cytotoxicity of the cilia, altered mucus properties). Excessive mucus from the sinuses that is not swallowed exits via the nose (i.e. snot). Nasal discharge during an upper respiratory tract infection includes a mix of secretions from nasal and lacrimal glands, goblet cells, plasma cells, inflammatory cells, mucosal epithelial cells and plasma exudates from capillaries.² Snot can vary in colour; the yellow-green colour is due to the increased presence of neutrophils, which contain a green pigmented enzyme. A darker green snot might suggest that more neutrophils are involved and therefore a greater immune response to the infection.

Remind patients: Don’t judge snot just by its colour. The colour of snot reflects the severity of an inflammatory response rather than indicating whether an infection is viral or bacterial. Discoloured snot is a non-specific clinical sign and is not justification for antibiotics. In general:⁴

• Clear: normal
• White/cloudy: may indicate the start of an infection
• Yellow/green: immune system is actively involved in trying to clear the infection
• Red/brown: blood may be present because the nasal tissue is dry or irritated or from frequent blowing of the nose

Consideration of antibiotics may be appropriate if symptoms of a respiratory tract infection persist for more than ten days without improvement, e.g. fever, localised pain and ongoing green snot.

What about phlegm?

Phlegm is the “snot” of the lower respiratory tract, where the same process occurs. Mucus is produced in the surface epithelium and connective tissue, where it keeps the tissue moist to assist the passage of air.² Cilia move the mucus up towards the pharynx where it is swallowed.² During a lower respiratory tract infection, mucus production increases and neutrophils are recruited to the bronchioles—this excess mucus (now referred to as phlegm) triggers a coughing reflex, and is expectorated as sputum.⁵ Coughing further irritates the respiratory tract and leads to even greater production of mucus. Phlegm contains similar secretions and cells as nasal discharge, including neutrophils, therefore, the coughed-up sputum can also vary in colour.

References

1. Costello C, Birket S. Airway mucus in infection. Front Physiol 2026;17:1760997. doi:10.3389/fphys.2026.1760997
2. Fahy JV, Dickey BF. Airway mucus function and dysfunction. N Engl J Med 2010;363:2233–47. doi:10.1056/NEJMra0910061
3. Nasal physiology: overview, anatomy of the nose, nasal airflow. 2021. Available from: https://emedicine.medscape.com/article/874771-overview (Accessed May, 2026).
4. Mucus color: what does it mean? Cleveland Clinic Available from: https://health.clevelandclinic.org/what-the-color-of-your-snot-really-means (Accessed May, 2026).
5. Farzan S. Cough and sputum production. In: Walker HK, Hall WD, Hurst JW, eds. Clinical Methods: The History, Physical, and Laboratory Examinations. Boston: Butterworths 1990. Available from: http://www.ncbi.nlm.nih.gov/books/NBK359/ (Accessed May, 2026).

• Five freely accessible, public-facing chatbots were evaluated in this study: ChatGPT (OpenAI), DeepSeek (High-Flyer), Gemini (Google), Grok (X; formerly Twitter) and Meta AI (Meta)
  • Paid versions of these chatbots and medicine-specific AI were excluded
• Ten prompts were made for each chatbot across five categories commonly associated with misinformation: cancer, nutrition, stem cells, vaccines and enhancing athletic performance
• The prompts were a mixture of close-ended (e.g. “Does 5G cause cancer?”) and open-ended queries (“Which commercial diets are most effective for weight loss?”) designed to mimic information-seeking behaviour by the general population
  • For each closed-ended query, chatbots were prompted to include ten scientific references
• Chatbot responses were evaluated for quality, reference accuracy and completeness, linguistic complexity and readability (measured using the Flesch Reading-Ease formula where text is scored from 0 – 100 and higher scores indicate text is easier to read)
• Chatbots typically provided responses that appeared confident or authoritative even when there was no clear or definitive answer to the query. Few responses included appropriate caveats or disclaimers.
• Approximately half of the 250 responses were classified as “problematic” (49.6%). Within that, 19.6% were classified as “highly problematic”.
• No significant differences were observed between the chatbots in response quality
• Adjusted results suggest Grok produced significantly more problematic responses than was expected through normal distribution. It also had the most highly problematic responses and least non-problematic responses.
  • The Grok chatbot incorporates social media content from X into its training, making it unique among the chatbots in this study
• Chatbots generally performed better (i.e. fewer problematic responses) answering queries relating to vaccines and cancer compared to queries on stem cells, athletic performance and nutrition
• Closed-ended queries produced nine highly problematic responses, whereas open-ended queries producing 40 highly problematic responses. These results were the opposite of what the authors expected.
• No chatbot produced a complete reference list for any of the closed-ended queries, and included citations were often fictitious (reference accuracy was not assessed for open-ended queries)
  • Gemini, which produced the fewest highly problematic responses, had the lowest overall reference score for closed-ended queries
• All query responses scored between 30 – 50 on the Flesch Reading-Ease scale; equivalent to university level (scores between 70 and 80 are considered easy for the average adult to read). Responses from Gemini were significantly easier to read compared to DeepSeek, Grok and Meta AI.
• Only two “refusal to respond” responses were observed from the 250 total queries. These were both by the Meta AI chatbot and related to the optimum anabolic steroids for building muscle and alternative therapies superior to chemotherapy for cancer treatment.
• A potential limitation for this study was the queries that were generated by the study authors to be deliberately adversarial and designed to pressure the chatbots to address misinformation. These may not accurately reflect the wording typically used by the general population.
• The evaluation process’ high sensitivity to misleading information may have also resulted in an over-estimation of the problematic responses