Validation of the treatment identification strategy of the HEDIS addiction quality measures: concordance with medical record review.

BACKGROUND: Strategies to accurately identify the occurrence of specific health care events in administrative data is central to many quality improvement and research efforts. Many health care quality measures have treatment identification strategies based on diagnosis and procedure codes - an approach that is inexpensive and feasible but usually of unknown validity. In this study, we examined if the diagnosis/procedure code combinations used in the 2006 HEDIS Initiation and Engagement quality measures to identify instances of addiction treatment have high concordance with documentation of addiction treatment in clinical progress notes. METHODS: Four type of records were randomly sampled from VHA electronic medical data: (a) Outpatient records from a substance use disorder (SUD) specialty clinic with a HEDIS-qualified substance use disorder (SUD) diagnosis/CPT code combination (n = 700), (b) Outpatient records from a non-SUD setting with a HEDIS-qualified SUD diagnosis/CPT code combination (n = 592), (c) Specialty SUD Inpatient/residential records that included a SUD diagnosis (n = 700), and (d) Non-SUD specialty Inpatient/residential records that included a SUD diagnosis (n = 700). Clinical progress notes for the sampled records were extracted and two raters classified each as documenting or not documenting addiction treatment. Rates of concordance between the HEDIS addiction treatment identification strategy and the raters' judgments were calculated for each record type. RESULTS: Within SUD outpatient clinics and SUD inpatient specialty units, 92% and 98% of sampled records had chart evidence of addiction treatment. Of outpatient encounters with a qualifying diagnosis/procedure code combination outside of SUD clinics, 63% had chart evidence of addiction treatment. Within non-SUD specialty inpatient units, only 46% of sampled records had chart evidence of addiction treatment. CONCLUSIONS: For records generated in SUD specialty settings, the HEDIS strategy of identifying SUD treatment with diagnosis and procedure codes has a high concordance with chart review. The concordance rate outside of SUD specialty settings is much lower and highly variable between facilities. Therefore, some patients may be counted as meeting the 2006 HEDIS Initiation and Engagement criteria without having received the specified amount (or any) addiction treatment.

Pharmacotherapy for alcohol dependence: perceived treatment barriers and action strategies among Veterans Health Administration service providers.

Although access to and consideration of pharmacological treatments for alcohol dependence are consensus standards of care, receipt of these medications by patients is generally rare and highly variable across treatment settings. The goal of the present project was to survey and interview the clinicians, managers, and pharmacists affiliated with addiction treatment programs within Veterans Health Administration (VHA) facilities to learn about their perceptions of barriers and facilitators regarding greater and more reliable consideration of pharmacological treatments for alcohol dependence. Fifty-nine participants from 19 high-adopting and 11 low-adopting facilities completed the survey (facility-level response rate = 50%) and 23 participated in a structured interview. The top 4 barriers to increased consideration and use of pharmacotherapy for alcohol dependence were consistent across high- and low-adopting facilities and included perceived low patient demand, pharmacy procedures or formulary restrictions, lack of provider skills or knowledge regarding pharmacotherapy for alcohol dependence, and lack of confidence in treatment effectiveness. Low patient demand was rated as the most important barrier for oral naltrexone and disulfiram, whereas pharmacy or formulary restrictions were rated as the most important barrier for acamprosate and extended-release naltrexone. The 4 strategies rated across low- and high-adopting facilities as most likely to facilitate consideration and use of pharmacotherapy for alcohol dependence were more education to patients about existing medications, more education to health care providers about medications, increased involvement of physicians in treatment for alcohol dependence, and more compelling research on existing medications. This knowledge provides a foundation for designing, deploying, and evaluating targeted implementation efforts.

Common statistical and research design problems in manuscripts submitted to high-impact medical journals.

BACKGROUND: To assist educators and researchers in improving the quality of medical research, we surveyed the editors and statistical reviewers of high-impact medical journals to ascertain the most frequent and critical statistical errors in submitted manuscripts. FINDINGS: The Editors-in-Chief and statistical reviewers of the 38 medical journals with the highest impact factor in the 2007 Science Journal Citation Report and the 2007 Social Science Journal Citation Report were invited to complete an online survey about the statistical and design problems they most frequently found in manuscripts. Content analysis of the responses identified major issues. Editors and statistical reviewers (n = 25) from 20 journals responded. Respondents described problems that we classified into two, broad themes: A. statistical and sampling issues and B. inadequate reporting clarity or completeness. Problems included in the first theme were (1) inappropriate or incomplete analysis, including violations of model assumptions and analysis errors, (2) uninformed use of propensity scores, (3) failing to account for clustering in data analysis, (4) improperly addressing missing data, and (5) power/sample size concerns. Issues subsumed under the second theme were (1) Inadequate description of the methods and analysis and (2) Misstatement of results, including undue emphasis on p-values and incorrect inferences and interpretations. CONCLUSIONS: The scientific quality of submitted manuscripts would increase if researchers addressed these common design, analytical, and reporting issues. Improving the application and presentation of quantitative methods in scholarly manuscripts is essential to advancing medical research.

Are VHA administrative location codes valid indicators of specialty substance use disorder treatment?

Healthcare quality managers and researchers often need to identify specific healthcare events from administrative data. In this study, we examined whether Veterans Health Administration (VHA) clinic stop and bed section codes are reliable indicators of substance use disorder (SUD) treatment as documented in clinical progress notes. For outpatient records with a progress note, SUD clinic stop code, SUD diagnosis code, and mental health procedure code, we found chart documentation of SUD care in 92.0% of 601 records: 82.5% of 372 records with a SUD clinic stop code and SUD diagnosis code but no mental health procedure code, 21.9% of 379 records with a SUD clinic stop code and mental health procedure code but no SUD diagnosis code, and 55.3% of 318 records with a SUD clinic stop code but no SUD diagnosis or mental health procedure code. For inpatient stays with a SUD bed section code and a progress note, we found chart documentation of SUD care in 99.0% of 699 records accompanied by a SUD diagnosis but 0% of 39 records without a SUD diagnosis. These results provide validity evidence and caveats to researchers and VHA quality managers who might use SUD specialty location codes as indicators of SUD specialty care.