Over a third of all Americans are at an increased risk for chronic conditions such as heart attack, stroke, cancer, and diabetes due to their lack of sleep. Key disorders, such as obstructive sleep apnea, which now afflicts 30 million adults in the United States, are some of the primary causes of the morbidity and mortality attributed to the lack of sleep. In this 2nd entry of the Bang for the Buck series, we investigate the possibility of investing in positive airway pressure (PAP), a mode of respiratory ventilation, sleep apnea’s first-line treatment as a way to reduce this morbidity.
Assessing the cost associated with sleep apnea is difficult because of interactions between outcomes. For example, untreated and undiagnosed obstructive sleep apnea is associated with some outcomes that are in turn related to coronary artery disease, hypertension, and diabetes. Furthermore, sleep disorders have been associated with increased incidence of occupational injuries and decreased work productivity.
Despite the difficulty, efforts have been made to assess the cost associated with sleep apnea. A white paper published by the American Academy of Sleep Medicine estimated that the cost may be near $150 billion for undiagnosed obstructive sleep apnea alone. The authors of the study included the costs of comorbidities, mental health, motor vehicle accidents, workplace accidents, and lost productivity. The study also found that diagnosing and treating patients cost $2,105 per patient, which is 33% of the non-treatment cost ($6,366). These findings suggest that diagnosing and treating individuals with obstructive sleep apnea may result in significant net savings for society.
Other studies have revealed PAP therapy to be cost-effective. A study on American drivers utilizing a state-transition Markov model to compare the costs and outcomes of obstructive sleep apnea treated with continuous positive airway pressure (CPAP) with no therapy for 5 years found that CPAP was cost-effective with an ICER of $3,354 per QALY gained. In another model-based study, CPAP was found to be cost-effective compared to no treatment after only 2 years.
Though these 2 studies are promising, they have limitations. For example, modeling studies are dependent on published data. With the added complexity of sleep apnea, limits of other studies adopted into the model may make it difficult to generalize the cost-effectiveness of PAP. A new study published ahead of print in Medical Care has attempted to fill in some of these gaps in empiric data. Using information from the Kaiser Permanente Southern California health system, the authors compared acute care utilization and medication utilization between diagnosed sleep apnea patients dispensed PAP therapy and propensity score-matched members not diagnosed with sleep apnea or dispensed PAP therapy. An index date was used to complete the negative binomial analyses used to compare the groups. The index date for those who received the therapy was the date PAP was dispensed. For the matched members the was a random date between the member’s enrollment and the case-identification period. The study used several years of data for the pre-period and post-period (up to five years before and seven years after).
The study’s authors hypothesized that those with PAP would have lower health care utilization as PAP therapy is expected to interrupt the generation of physiologic stressors. The study’s conceptual framework explains that because stressors associated with sleep apnea contribute to diseases and adverse health behaviors, not receiving PAP would lead to increased acute care utilization and use of medication.
However, the results did not turn out as expected. Interestingly, trends in the annual rate of health care utilization were not significantly different for the groups. The change in acute care utilization pre-index to post-index of the group dispensed with PAP compared to the change in the matched group reveals a risk ratio of 0.98, or a mere 2% reduction in risk. Medication utilization, on the other hand, had an even smaller risk ratio of 1.008. This indicates that again the change pre-index to post index date in those dispensed with PAP was similar to the trend in the matched group with only a 0.8% increase in risk. These similarities in trends lead the authors to conclude that the PAP dispensation observed in the study does not reduce the annual rate of acute care and medication utilization.
The level of adherence could explain why PAP dispensation did not reduce the annual rate of healthcare utilization. A review regarding nights per week of CPAP non-use reveals the percentages range from 10% to 40%. An argument could be made that though CPAP is cost-effective, low adherence negates the positive effects of this technology. Like tobacco dependence treatment service discussed in the first Bang for the Buck, we are again reminded of how promotion and education can play a vital role in our investment in healthcare technology. As the king of advertising, Phineas Taylor Barnum, once said, “Without promotion, something terrible happens… nothing!”
See you in the next Bang for the Buck!
Red Thaddeus Miguel