A Corporate Driven Sleep Apnea Detection and Treatment Program: Results and Challenges
Page 3 of 5: Results (Continued...)
Screening Tool
Analysis of screening tool data is available from only one testing source (PPD). A total of 62 drivers were referred to this source for testing. Of those 62 individuals, data was incomplete for 5, leaving 57 drivers for this analysis. Of these 57, 49 are males and 8 are females. Because of the small number of females, they were not separately analyzed from the males. The sensitivity of our screening tool to detect SDB of any severity was 86% and remained at 86% if used to detect only moderate to severe SDB. The specificity was 43% in all SDB and only 24% in moderate to severe SDB. Overall, the accuracy was 80%. The positive predictive value was 91% in predicting the presence of any severity of SDB and fell to 72% for moderate and severe SDB. The negative predictive value was only 30% for all severity SDB and improved only to 40% in moderate and severe SDB. Twenty-five of the 57 drivers (44%) analyzed in this population had severe SDB with apnea/hyponea scores ranging from 34 to 112 events per hour of sleep.
Health Costs
The number of study group members analyzed by time period is shown below in Table 1:
Table 1: Number of Study Group Members Receiving CPAP Intervention by Time Period
| |
2003 |
2004 |
Jan-June 2005 |
| Number of Members |
3 |
257 |
88 |
Table 2 illustrates the breakdown of health care costs generated from enrolled members with SDB before and after CPAP intervention. Overall, there was a 47.8% reduction in health care costs after CPAP intervention ( p<.0001). The mean values for admissions PMPM and outpatient visits PMPM are significantly less in the post-CPAP period than in the pre-CPAP period( p<.001 and p<.0001, respectively). Medical spending also decreased significantly ( p<.0001). There was a modest increase in prescription spending (mean difference of $32.84 per month), along with a 42.6% increase in PMPM office visits. After CPAP intervention, affected SDB patients experienced roughly one fourth as many hospital admissions and spent about one half the health care dollars compared with the period prior to intervention.
Table 2: Summary Statistics and T-Test Results for CPAP Intervention (n=348)
| |
Pre-Intervention |
Post-Intervention |
% Change |
| Admits PMPM |
|
|
|
| Mean |
.0128 |
.0035 |
-72.7% ** |
| Std. Dev. |
.0432 |
.0242 |
|
| ER Visits PMPM |
|
|
|
| Mean |
.0224 |
.0238 |
6.3% |
| Std. Dev. |
.0676 |
.0666 |
|
| Office Visits PMPM |
|
|
|
| Mean |
.7621 |
1.087 |
42.6% *** |
| Std. Dev. |
.7869 |
1.239 |
|
| OP Visits PMPM |
|
|
|
| Mean |
.2663 |
.1245 |
-53.2% *** |
| Std. Dev. |
.3954 |
.2418 |
|
| Allowed Amount Medical PMPM |
|
|
|
| Mean |
$807.58 |
$341.15 |
-57.8% *** |
| Std. Dev. |
$1,294.80 |
$553.59 |
|
| Allowed Amount Rx PMPM |
|
|
|
| Mean |
$98.70 |
$131.54 |
33.3% * |
| Std. Dev. |
$154.40 |
$267.27 |
|
| Allowed Amount Total PMPM |
|
|
|
| Mean |
$906.28 |
$472.69 |
-47.8% *** |
| Std. Dev. |
$1,320.45 |
$653.92 |
|
* p < .05
** p < .001
*** p < .0001
CPAP Compliance
Forty-six drivers diagnosed and treated for sleep apnea with CPAP for at least one month (range 1 to 9 month follow-up) have been contacted with 32 drivers responding (response rate of 69%). Of those drivers responding, 91% are using CPAP 6-7 nights per week. Most are very pleased with their treatment, giving a subjective average score of 7.5 and a mean score of 8.5 out of a possible 10.
Accident Rates
Of the drivers diagnosed and treated for SDB and eligible for analysis (see Methods section above for exclusion criteria), 75%, had incurred a preventable accident during this analysis. Of those drivers with accidents, 93% had their accident before treatment for SDB. Twenty-five percent incurred an accident after treatment for SDB. Of those drivers with accidents and diagnosed SDB, only 7% incurred an accident after treatment for SDB, but not before. Overall, the accident rate in these treated individuals dropped from 93% to 25%, or a 73% reduction.
Corporate Retention Rates
Extremely high driver turn-over is prevalent throughout the industry. Industry-wide norms in yearly retention rates are as low as 25%. Although SNI has asked that their yearly retention rates remain proprietary and not be reported, our survey found that SNI SDB-treated drivers had a 129% greater retention rate (odds ratio 2.29) compared to 2004 SNI global driver retention rates.
