Mentation Tracks Severity but not Oxygenation in Obstructive Sleep Apnea.

There is a rough consensus, after decades of research, that obstructive sleep apnea (OSA) is associated with mild cognitive impairments, especially in areas of executive functioning (EF), attention/working memory (A/WM), episodic memory (EM), and speed of speed of information processing (SIP). However, there is less consensus as to whether apnea severity matters for these impairments, which sleep variables matter most to which cognitive domains, whether common OSA comorbidities contribute to these determinations, or whether the apparent associations are really artifacts of these comorbidities. In this study, 40 participants with OSA submitted to polysomnography and to neuropsychological assessment with an expanded Halstead-Reitan Test Battery. Aggregates of tests to cover the four cognitive domains mentioned above were linearly regressed on the apnea-hypopnea index (AHI), the nadir of oxygen saturation (NOS), and hypertension and diabetes mellitus (scored present or absent). The AHI predicted both EF (p = .015; sr2 = .13) and A/WM (p = .023; sr2 = 11) in the primary analyses, and EM (p = .027; sr2 = .10) in the secondary analyses. Thus, AHI may affect EF, A/WM and perhaps EM beyond NOS and beyond two of OSA's most common comorbidities. Implications of these findings are discussed here.

Construct and incremental validity of the Rotter Incomplete Sentences Blank in adult psychiatric outpatients.

The expression of positive vs negative attitudes in response to the 40 stems of the Rotter Incomplete Sentences Blank (RISB) can be scored and summed to an Overall Adjustment Score (OAS). By extending the validation of the OAS to adult psychiatric outpatients, this study demonstrates for the first time the incremental validity of a personality test over a simple self-rating. The RISB and six tests of adjustment were administered to 41 recent admissions to psychotherapy in two rural clinics. The tests of adjustment were selected to cover the two domains of symptoms vs function and the three methods of interview schedule vs objective inventory vs therapist ratings. Their scores were combined into an adjustment composite. The OAS related strongly to the composite in univariate regression, and moderately in hierarchical regression after covarying demographics, intelligence, social desirability, and self-ratings of adjustment. Construct validity was good, and incremental validity at least fair.