Hospital-Acquired DVT/PE As “Never Events”: A Misguided Strategy for Performance Improvement in VTE Prevention.

Venous thromboembolism (VTE) is recognized as a common cause of preventable patient harm. To reduce the incidence of preventable VTE in hospitalized patients, VTE has been designated as a “Never Event”, a complication for which reimbursement is withheld. To identify VTE “Never Events” that occur during the hospital stay, payors are using the International Classification of Disease revision 9 (ICD-9) diagnosis codes for VTE. The estimated positive predictive value of ICD-9 diagnosis codes for objectively confirmed acute hospital-acquired VTE varies widely (17% to 100%). The purpose of the current study was to review consecutive cases of hospital-acquired VTE identified during fiscal year 2011 by the Maryland Hospital Acquired Conditions (MHAC) algorithm for VTE “Never Events.” We hypothesized that a significant number of VTE identified using this strategy would not fulfill the definition of preventable episodes of VTE.

We conducted a retrospective review of an administrative database of consecutive patients admitted to the Johns Hopkins Hospital during fiscal year 2011 who suffered events coded as hospital-acquired VTE. The ICD-9 codes used by the MHAC initiative to define VTE are: 451.11, 451.19, 451.2, 451.81, 451.83, 451.9, 453.40, 453.41, 453.42, 453.82, 453.84, 453.85, 453.86, 453.87, 453.89, 453.9, and 997.2. Demographic and clinical data including VTE risk category and ordered VTE prophylaxis were retrieved from our electronic patient record system. Our definition for a hospital-acquired VTE required objective radiologic confirmation of lower extremity deep venous thrombosis (DVT) or pulmonary embolism (PE) in the absence of a central venous catheter (CVC). Thrombotic events were consider CVC-DVT if they occurred in the same vascular distribution. VTE were considered present on admission if objectively-confirmed within the first 24 hours of admission.

Between July 1, 2010 and June 30, 2011, one hundred sixty three patients were coded and accepted as developing preventable VTE by the MHAC initiative. Ninety one patients were admitted to a surgical service, 58 patients were medically ill, and 14 were neurology patients. The mean age was 61.2 (SD: 17.5) years, mean BMI was 28.0 (SD: 7.6) kg/m², median length of hospitalization was 14 days and median time to diagnosis was 5 days. One hundred four patients (63.8%) were categorized as being at high or very high risk for VTE while 59 patients (36.2%) were determined to be at moderate risk for VTE.

Eighty seven patients (53.4%) had DVT, 47 (28.8%) had PE, 9 (5.5%) had superficial thrombophlebitis, and 19 (11.7%) had arterial thromboembolism. One patient, who died on post-operative day 1 without an objective finding of VTE, was coded as developing VTE. Fifty eight patients (35.6%) had CVC-associated events and 7 patients (4.3%) had VTE that were present on admission. Overall, 85 of 163 (52%) thrombotic events designated as VTE “Never Events” were in fact either not preventable (CVC-VTE) or not hospital-acquired VTE.

Our retrospective study of consecutive patients designated as having developed hospital-acquired VTE found that a substantial percentage of VTE “Never Events” are invalid or unlikely to be preventable with currently available best practices VTE prophylaxis. These findings highlight the inaccuracy of ICD-9 code based strategies to identify potentially preventable hospital-acquired conditions. Furthermore, the data emphasize that process measures rather than outcomes should be used to determine performance on quality indicators such as VTE prevention. We believe that a method with greater sensitivity and specificity to identify true VTE is needed, which may be aided by ICD-10-CM. While we agree with the concept of disincentivizing suboptimal care, a more practical approach may be to evaluate process measures, such as prescribing and administering guideline-based, best-practice VTE prophylaxis for patients who develop VTE.

Streiff:sanofi-aventis: Consultancy, Honoraria; BristolMyersSquibb: Research Funding; Eisai: Consultancy; Janssen Healthcare: Consultancy; Daiichi-Sankyo: Consultancy.