Positive Predictive Value of Administrative Data for Neonatal Abstinence Syndrome.

OBJECTIVES: Neonatal abstinence syndrome (NAS) is a postnatal withdrawal syndrome experienced by some infants with opioid exposure. Hospital administrative data are commonly used for research and surveillance but have not been validated for NAS. Our objectives for this study were to validate the diagnostic codes for NAS and to develop an algorithm to optimize identification. METHODS: Tennessee Medicaid claims from 2009 to 2011 (primary sample) and 2016 (secondary sample; post-International Classification of Diseases, 10th Revision, Clinical Modification [ICD-10-CM]) were obtained. Cases of NAS were identified by using International Classification of Diseases, Ninth Revision, Clinical Modification code (2009-2011) 779.5 and ICD-10-CM code (2016) P96.1. Medical record review cases were then conducted by 2 physicians using a standardized algorithm, and positive predictive value (PPV) was calculated. Algorithms were developed for optimizing the identification of NAS in administrative data. RESULTS: In our primary sample of 112 029 mother-infant dyads, 950 potential NAS cases were identified from Medicaid claims data and reviewed. Among reviewed records, 863 were confirmed as having NAS (including 628 [66.1%] cases identified as NAS requiring pharmacotherapy, 224 [23.5%] as NAS not requiring pharmacotherapy, and 11 [1.2%] as iatrogenic NAS), and 87 (9.2%) did not meet clinical criteria for NAS. The PPV of the International Classification of Diseases, Ninth Revision, Clinical Modification code for NAS in clinically confirmed NAS was 91% (95% confidence interval: 88.8%-92.5%). Similarly, the PPV for the ICD-10-CM code in the secondary sample was 98.2% (95% confidence interval: 95.4%-99.2%). Algorithms using elements from the Medicaid claims and from length of stay improved PPV. CONCLUSIONS: In a large population-based cohort of Medicaid participants, hospital administrative data had a high PPV in identifying cases of clinically diagnosed NAS.

Evaluation and disposition of Medicaid-insured children and adolescents with suicide attempts.

OBJECTIVE: Guidelines and quality of care measures for the evaluation of adolescent suicidal behavior recommend prompt mental health evaluation, hospitalization of high-risk youth, and specific follow-up plans-all of which may be influenced by sociodemographic factors. The aim of this study was to identify sociodemographic characteristics associated with variations in the evaluation of youth with suicidal behavior. METHODS: We conducted a large cohort study of youth, aged 7 to 18, enrolled in Tennessee Medicaid from 1995 to 2006, who filled prescriptions for antidepressants and who presented for evaluation of injuries that were determined to be suicidal on the basis of external cause-of-injury codes (E codes) and ICD-9-CM codes and review of individual medical records. Chi-square tests and logistic regression were performed to assess the relationship between sociodemographic characteristics and documentation of mental health evaluation, hospitalization, and discharge instructions. RESULTS: Of 929 episodes of suicidal behavior evaluated in an acute setting, rural-residing youth were less likely to be admitted to a psychiatric hospital (adjusted odds ratio [AOR] 0.72; 95% confidence interval [CI] 0.55-0.95) and more likely to be medically hospitalized only (AOR 1.92; 95% CI 1.39-2.65). Female subjects were less likely to be admitted to a psychiatric hospital (AOR 0.55; 95% CI 0.41-0.74) and more likely to be discharged home (AOR 1.44; 95% CI 1.01-2.04). Only 40% of those discharged to home had documentation of discharge instructions with both follow-up provider and date. CONCLUSIONS: In this statewide cohort of youth with suicidal behavior, there were significant differences in disposition associated with sociodemographic characteristics.

Identifying suicidal behavior among adolescents using administrative claims data.

PURPOSE: To assess the safety of psychotropic medication use in children and adolescents, it is critical to be able to identify suicidal behaviors from medical claims data and distinguish them from other injuries. The purpose of this study was to develop an algorithm using administrative claims data to identify medically treated suicidal behavior in a cohort of children and adolescents. METHODS: The cohort included 80,183 youth (6-18 years) enrolled in Tennessee's Medicaid program from 1995-2006 who were prescribed antidepressants. Potential episodes of suicidal behavior were identified using external cause-of-injury codes (E-codes) and ICD-9-CM codes corresponding to the potential mechanisms of or injuries resulting from suicidal behavior. For each identified episode, medical records were reviewed to determine if the injury was self-inflicted and if intent to die was explicitly stated or could be inferred. RESULTS: Medical records were reviewed for 2676 episodes of potential self-harm identified through claims data. Among 1162 episodes that were classified as suicidal behavior, 1117 (96%) had a claim for suicide and self-inflicted injury, poisoning by drugs, or both. The positive predictive value of code groups to predict suicidal behavior ranged from 0-88% and improved when there was a concomitant hospitalization but with the limitation of excluding some episodes of confirmed suicidal behavior. CONCLUSIONS: Nearly all episodes of confirmed suicidal behavior in this cohort of youth included an ICD-9-CM code for suicide or poisoning by drugs. An algorithm combining these ICD-9-CM codes and hospital stay greatly improved the positive predictive value for identifying medically treated suicidal behavior.

An algorithm to identify incident myocardial infarction using Medicaid data.

PURPOSE: Studies of non-steroidal anti-inflammatory drugs (NSAIDs) and cardiovascular events using administrative data require identification of incident acute myocardial infarctions (AMIs) and information on whether confounders differ by NSAID status. METHODS: We identified patients with a first AMI hospitalization from Tennessee Medicaid files as those with primary ICD-9 discharge diagnosis 410.x and hospitalization stay of > 2 calendar days. Eligible persons were non-institutionalized, aged 50-84 years between 1999-2004, had continuous enrollment and no AMI, stroke, or non-cardiovascular serious medical illness in the prior year. Of 5524 patients with a potential first AMI, a systematic sample (n = 350) was selected for review. Using defined criteria, we classified events using chest pain history, EKG, and cardiac enzymes, and calculated the positive predictive value (PPV) for definite or probable AMI. RESULTS: 337 of 350 (96.3%) charts were abstracted and 307 (91.1%), 6 (1.8%), and 24 (7.1%) events were categorized as definite, probable, and no AMI, respectively. PPV for any definite or probable AMI was 92.8% (95% CI 89.6-95.2); for an AMI without an event in the past year 91.7% (95% CI 88.3-94.2), and for an incident AMI was 72.7% (95% CI 67.7-77.2). Age-adjusted prevalence of current smoking (46.4% vs. 39.1%, p = 0.35) and aspirin use (36.9% vs. 35.9%, p = 0.90) was similar among NSAID users and non-users CONCLUSIONS: ICD-9 code 410.x had high predictive value for identifying AMI. Among those with AMI, smoking and aspirin use was similar in NSAID exposure groups, suggesting these factors will not confound the relationship between NSAIDs and cardiovascular outcomes.

Snoddy (1926) revisited: time scales of motor learning.

The authors investigated the time scales of the learning of a mirror-tracing task to reexamine G. S. Snoddy's (1926) original claim and the received theoretical view (A. Newell & P. S. Rosenbloom, 1981) that motor learning follows a power law. Adult participants (N = 16) learned the tracing task in either a normal or a reversed visual-image condition over 5 consecutive days of practice and then performed 1 day of practice 1 week later and again 1 month later. The reversed-image group's performance was poorer than that of the normal-image group throughout the practice. An exponential was the best fitting function on individual data, but the power-law function was the best fit on the group-averaged data. The findings provided preliminary evidence that 2 characteristic time scales, (a) fast, dominated by warm-up, and (b) slow, dominated by persistent change, capture individuals' performance in the learning of the mirror-tracing task.