However, the others suffering from the severe OSA would not have been found and many with AHI variability would have been missed. . Ahmadi et al. (2009) wanted to investigate AHI on two nights due to a concern with variability of the disorder and impact on clinical diagnosis. They conducted polysomnographies with 193 sleep clinic patients over two consecutive nights to analyze AHI variability. Anonymized records from five individuals with significant night-to-night AHI variability participated: the two-night tests from two patients were represented as four individual polysomnographies; the two-night tests for two others were represented as being obtained from two different sleep clinics; the last patient's results were shown as a two-night study. Twenty-two sleep experts at the Associated Professional Sleep Societies meeting diagnosed the results after being told that tests were from seven patients: four with single-night and two with two polysomnographies, each one from a different clinic; and one patient with a two-night test. The results showed that 21% of the patients had a nightly AHI variability of more than 5. Forty-eight percent of everyone tested had a significantly higher AHI on the first night, and 41% had a significantly higher AHI on the second night. Sleep apnea due to low AHI would have been undetected in 20% of the people on one night. In addition, 13% of all patients had a more severe sleep apnea classification based on the second night of polysomnographies. In regard to the seven cases, 27% to 36% of sleep experts did not accurately identify sleep apnea especially when presented with the polysomnographies containing the lower AHI. Frequency of missed sleep apnea diagnoses were reduced to 15% to 18% when information from two tests were presented to the sleep professionals. This study reinforces the variability that exists from one night to the next in polysomnography respiratory testing. In addition, identification of sleep apnea in some individuals is reduced when sleep experts are provided with only one recording result. The clinical implication is that approximately13% of sleep clinic patients may benefit from a second night of testing.

As noted previously, at-home testing normally have to be confirmed. Levendowski et al. (2009) tested the variability of AHI obtained in home at four- to six-month intervals and through polysomnography....

When comparing the test-retest AHI and AI, the in-home results were more highly correlated than the comparable polysomnographic results. The in-home results provided about 50% less test-retest variability than the comparable polysomnography AHI and AI values. The authors found that in-home studies provided a repeated measure of sleep disordered breathing less variable then polysomnography. They recommend that experts using polysomnography to assess treatment outcomes need to factor in the increased variability and bias toward increased AHI values upon retest to ensure the accuracy of the study.
The field of polysomnography continues to grow, with more colleges and universities adding the program to their studies. There were three accredited labs in 1977 and over 1,800 American Academy of Sleep Medicine accredited labs and centers in the U.S. In 2008.