Three quarters of patients with recalled Philips CPAP devices reduced their nightly usage after the 2021 TGA notice, even as regulators urged them to keep using the machines.
A new single-centre study from Austin Health found mean nightly CPAP usage fell by 1.33 hours in the six months after the Therapeutic Goods Administration issued its product defect correction notice on 2 July 2021, dropping from 5.56 to 4.23 hours per night (95% CI 1.11, 1.55; P<0.001). The effect size was large (Cohen’s d = 2.14).
Of 368 patients with usable before-and-after data, 76.1% decreased usage, 23% increased it, 6.1% of those who decreased use stopped altogether, and just 1.1% made no change.
The TGA had warned that “sudden cessation of treatment could have an immediate detrimental effect on patient health,” and advised patients to consult their doctors before making any changes. It clearly did not reach many of them.
The authors, writing in the Internal Medicine Journal [link here] noted that “media coverage of the recall was often alarmist and there was a delay in access to healthcare services in the setting of a global pandemic,” making it likely many patients acted on publicly available information rather than medical advice.
The Philips recall, announced 14 June 2021, affected three to four million devices globally. The trigger was degradation of polyester-based polyurethane (PE-PUR) foam in some models, which could release particulate matter and volatile organic compounds when inhaled.
The Austin Health cohort was drawn from its subsidised CPAP programme for financially disadvantaged patients, which exclusively used Philips devices at the time. None were in a position to buy or hire a replacement.
Follow-up phone calls painted a picture of widespread patient alarm. “Many participants expressed significant anxiety about the potential health risks associated with recalled devices, with some choosing to discontinue use as a precaution,” the authors wrote. At the same time, “many participants expressed preference for advice in consultation with a clinician,” suggesting unmet demand for specialist guidance during the upheaval.
The patient cohort had severe OSA on average, with a mean Apnoea-Hypopnoea Index of 47.7. While usage at both time points remained above the four-hour threshold generally considered adequate for therapeutic efficacy, the authors noted that evidence points to a dose-dependent relationship between hours of use and clinical benefit.
Patients who reduced or stopped use tended to be older with lower baseline BMI and AHI, consistent with milder disease and a lower perceived risk from interrupting treatment. Those who increased their usage were younger with higher BMI and AHI, suggesting a greater perceived need to maintain therapy despite the safety concerns.
“Many participants with recalled devices did not follow the national TGA guidelines to continue usage of their recalled device as per prior usage habits, suggesting that concerns over potential health risks, limited access to timely medical advice and other external factors played a role in this decline,” the authors concluded.
They called for better strategies to disseminate safety information rapidly during large-scale device disruptions, saying it was “imperative to optimise strategies to provide timely, accessible medical information to participants to mitigate the adverse effects of suboptimal OSA treatment.”
The study was led by Dr Kirra Liu from the Department of Respiratory and Sleep Medicine at Austin Health.