Abstract
Background and Purpose:
The significance of transient neurological attack (TNA) symptoms is unclear. We sought to determine the risk of ischemic stroke after discharge from the emergency department (ED) with a diagnosis consistent with symptoms of TNA.
Methods:
Using administrative claims data, we identified patients discharged from EDs in New York between 2006 and 2012 with a primary discharge diagnosis of a TNA symptom, defined as altered mental status, generalized weakness, and sensory changes. The primary outcome was ischemic stroke. We used Kaplan-Meier survival statistics to calculate cumulative rates, and Cox regression to compare stroke risk after TNA versus after transient ischemic attack (TIA; positive control) or renal colic (negative control) while adjusting for demographics and vascular risk factors.
Results:
Of 499 369 patients diagnosed with a TNA symptom and discharged from the ED, 7756 were hospitalized for ischemic stroke over a period of 4.7 (±1.9) years. At 90 days, the cumulative stroke rate was 0.29% (95% confidence interval [CI]: 0.28%-0.31%) after TNA symptoms versus 2.08% (95% CI: 1.89%-2.28%) after TIA and 0.03% (95% CI: 0.02%-0.04%) after renal colic. The hazard ratio (HR) of stroke was higher after TNA than after renal colic (HR: 2.13; 95% CI: 1.90-2.40) but significantly lower than after TIA (HR: 0.47; 95% CI: 0.44-0.50). Compared to TIA, TNA was less strongly associated with stroke among patients under 60 years of age compared to those over 60.
Conclusions:
Patients discharged from the ED with TNA symptoms faced a higher risk of ischemic stroke than patients with renal colic, but the magnitude of stroke risk was low, particularly compared to TIA.
Keywords: stroke, cerebrovascular disorders, transient ischemic attack
Introduction
Transient neurological symptoms are termed transient ischemic attack (TIA) when the symptoms have a clear vascular anatomical localization. Patients with TIA face a high risk of subsequent ischemic stroke, particularly in the short-term.1,2 Rapid evaluation and management of patients with TIA has been associated with a reduction in stroke risk.3,4
Patients frequently present to emergency departments (EDs) with nonspecific neurological symptoms that resist a clear etiologic diagnosis. These symptoms, variably defined, have been termed “non-focal transient neurological attacks (TNAs)” and “nonspecific neurological attacks.”5,6 Such symptoms have previously been associated with only a slightly increased risk of stroke.5 On the other hand, a recent study found evidence of acute cerebral infarction on brain magnetic resonance imaging (MRI) in nearly 25% of a cohort of patients diagnosed with TNA by stroke neurologists at a specialized referral clinic.7 An additional study of patients evaluated at a specialized TIA referral clinic found that 10% of patients with recent isolated atypical TIA symptoms had a final diagnosis, upon imaging, of minor stroke.8 Given these conflicting findings, it remains unclear whether transient neurological symptoms not classified as TIA represent benign entities or portend a significant stroke risk.
In this study, we sought to understand the population-level clinical relevance of these types of transient neurological symptoms by investigating the risk of stroke in patients discharged from the ED with a primary diagnosis of a TNA symptom.
Methods
Design
We performed a retrospective cohort study using administrative claims data from all nonfederal ED visits in New York State from 2006 through 2013. The New York State Department of Health Statewide Planning and Research Cooperative System provides standardized discharge data to the Agency for Healthcare Research and Quality, which assigns each patient an anonymous identifier to facilitate de-identified, longitudinal tracking.9 The percentage of patients with a verified identifier during the years of our analysis was consistently greater than 94% for adult patients.10 These data include demographic information, an exhaustive list of International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis codes, and an indicator for radiology services such as MRI. The Weill Cornell Medicine institutional review board approved our analysis of these data and waived the need for informed consent.
Patient Population
We identified a cohort of all patients discharged from the ED with an ICD-9-CM code for a nonetiologic, symptom-based, neurological diagnosis in the primary discharge diagnosis position. Prior to any analysis, we exhaustively searched all ICD-9-CM codes for nonspecific neurological symptoms that appeared compatible with symptoms included in published definitions of nonfocal TNA (Table 1).5,7 Symptoms of focal TNA, which are symptoms recognized as having a clear vascular anatomic localization, were not included.5 As positive controls, we also included patients discharged from the ED with a primary discharge diagnosis of TIA (ICD-9-CM code 435.x), since TIA is a known herald of stroke. This diagnosis code for TIA has been shown to have positive predictive values of 70% or greater.11 Following the methods of other studies on this topic,12 we selected patients discharged with a primary discharge diagnosis of renal colic (ICD-9-CM code 788.0) as negative controls. Patients with a documented stroke prior to or at their index visit were excluded. Nonresidents of New York State were excluded to maximize follow-up. Since we had data through 2013, we excluded patients with an index visit in 2013 so as to have at least 1 year of follow-up for all patients.
Table 1.
Transient Neurological Attack Symptoms as Defined by International Classification of Diseases, Ninth Revision, Clinical Modification Codes.
| Symptom Description | Diagnosis Code(s) |
|---|---|
| Alteration of consciousnessa | 780.01, 780.09 |
| Altered mental statusa | 780.97 |
| Dizziness and giddinessa | 780.4 |
| Memory lossa | 780.93 |
| Malaise and fatigue | 780.79 |
| Nervousness, “nerves” | 799.2 |
| Acute confusional statea | 293.0 |
| Blurred vision, not otherwise specified | 368.8 |
| Subjective visual disturbance | 368.1x |
| Disturbance of sensation of smell and taste | 781.1 |
| Muscle weakness (generalized) | 728.87 |
| Paralysis, unspecified | 344.9 |
| Abnormal involuntary movements | 781.0 |
| Cramp | 729.82 |
| Spasm of muscle | 728.85 |
| Abnormal skin sensationb | 782.0 |
aIncluded as altered mental status in secondary analysis.
bThis includes anesthesia of skin, burning/prickling sensation, hyperesthesia, hypoesthesia, numbness, paresthesia, and tingling.
Measurements
The primary outcome was any subsequent hospital admission for acute ischemic stroke, as defined by a validated ICD-9-CM discharge diagnosis code algorithm.13 This algorithm has 95% specificity and 88% positive predictive value for ischemic stroke. Patients were censored at the time of in-hospital death or December 31, 2013. Using ICD-9-CM codes, we identified the following vascular risk factors and comorbidities at the time of the index visit: hypertension, diabetes, coronary artery disease, congestive heart failure, atrial fibrillation, peripheral vascular disease, chronic obstructive pulmonary disease, chronic kidney disease, tobacco use, and alcohol abuse.
Data Analysis
We used standard descriptive statistics with exact confidence intervals (CIs) to report crude rates. Kaplan-Meier survival statistics were used to calculate incidence rates and cumulative rates of stroke. We used Cox proportional hazards modeling to assess the relationship between the index diagnosis (ie, TNA, TIA, or renal colic) and subsequent stroke, while adjusting for demographics and vascular risk factors and comorbidities. In a subgroup analysis, we stratified patients by age (under 60 and greater than or equal to 60 years of age) in accordance with the ABCD2 score TIA risk stratification model.14 We also performed several sensitivity analyses. First, we limited our TNA cohort to only patients discharged from the ED with primary discharge diagnoses of encephalopathy or nonspecific sensory symptoms, because these are the 2 symptoms that appeared to be most strongly associated with acute infarction on MRI.7 Second, we limited our cohort to those patients discharged from the ED without having undergone either computed tomography (CT) or MRI, because we were interested in whether such patients with limited evaluation represented an especially at-risk population. Third, in a post hoc sensitivity analysis, to account for the association between hospital-based diagnosis of migraine and subsequent stroke,15 we included prior hospital-based migraine (ICD-9-CM code 346) as an additional covariate in the Cox proportional hazards model. This diagnosis code has been used to ascertain migraine diagnoses in prior population-based studies.16 The proportional hazards assumption was confirmed by the inspection of log–log plots. Missing data were not imputed. All analyses were performed using Stata/MP, version 13 (StataCorp, College Station, Texas). The threshold of statistical significance allowed for an α error of .05.
Results
We identified 499 369 patients discharged from the ED with a primary discharge diagnosis code consistent with TNA symptoms. We also identified 78 273 patients with renal colic and 20 445 patients with TIA. The mean age of patients with TNA was 51.0 (±19.4) years, and 62.3% were female. Patients with TIA were older and had more vascular risk factors than patients with TNA symptoms, who were in turn older and had more vascular risk factors than patients with renal colic (Table 2). Prior to discharge, 185 957 (37.2%) patients with TNA symptoms underwent CT, 12 977 (2.6%) underwent MRI, and 260 782 (52.2%) underwent electrocardiogram; these rates were substantially lower than for patients discharged with TIA (Table 2).
Table 2.
Characteristics of Patients Stratified by Exposure Group.
| Characteristica | Transient Neurological Attack Symptom, N = 499 369 | Transient Ischemic Attack, N = 20 445 | Renal Colic, N = 78 273 |
|---|---|---|---|
| Age, mean (SD), years | 51.0 (19.4) | 68.8 (15.9) | 44.0 (14.5) |
| Female | 310 846 (62.3) | 12 019 (58.8) | 30 598 (39.1) |
| Raceb | |||
| White | 244 636 (49.5) | 14 946 (74.0) | 49 566 (64.0) |
| Black | 108 925 (22.1) | 2589 (12.8) | 6256 (8.1) |
| Hispanic | 80 444 (16.3) | 1586 (7.9) | 12 983 (16.8) |
| Asian | 14 754 (3.0) | 271 (1.3) | 2245 (2.9) |
| Other | 45 284 (9.2) | 811 (4.0) | 6362 (8.2) |
| Payment sourcec | |||
| Medicare | 111 044 (22.2) | 11 071 (54.2) | 6380 (8.2) |
| Medicaid | 67 722 (13.6) | 1346 (6.6) | 8171 (10.4) |
| Commercial | 241 499 (48.4) | 6850 (33.5) | 49 436 (63.2) |
| Self-pay | 68 766 (13.8) | 908 (4.4) | 13 149 (16.8) |
| Other | 10 334 (2.1) | 270 (1.3) | 1136 (1.5) |
| Income quartiled | |||
| 1 | 165 739 (33.5) | 4765 (23.9) | 17 123 (22.1) |
| 2 | 123 219 (24.9) | 5918 (29.7) | 18 222 (23.5) |
| 3 | 109 811 (22.2) | 5213 (26.1) | 19 234 (24.8) |
| 4 | 95 880 (19.4) | 4057 (20.3) | 22 925 (29.6) |
| Hypertension | 105 362 (21.1) | 11 057 (54.1) | 7531 (9.6) |
| Diabetes | 43 724 (8.8) | 4081 (20.0) | 3704 (4.7) |
| Congestive heart failure | 4250 (0.9) | 773 (3.8) | 81 (0.1) |
| Coronary heart disease | 18 137 (3.6) | 3296 (16.1) | 980 (1.3) |
| Peripheral vascular disease | 1687 (0.3) | 487 (2.4) | 109 (0.1) |
| COPD | 6963 (1.4) | 976 (4.8) | 275 (0.4) |
| Chronic kidney disease | 4013 (0.8) | 694 (3.4) | 138 (0.2) |
| Atrial fibrillation | 6736 (1.4) | 1572 (7.7) | 198 (0.3) |
| Tobacco use | 3261 (0.7) | 760 (3.7) | 208 (0.3) |
| Alcohol abuse | 19 632 (3.9) | 1387 (6.8) | 2068 (2.6) |
| Computed tomography | 185 957 (37.2) | 18 730 (91.6) | NA |
| Magnetic resonance imaging | 12 977 (2.6) | 5308 (26.0) | NA |
| Electrocardiogram | 260 782 (52.2) | 17 474 (85.5) | 5490 (7.0) |
Abbreviations: COPD, chronic obstructive pulmonary disease; SD, standard deviation.
aData are presented as number (%) unless otherwise specified.
bSelf-reported by patients or their surrogates. Numbers do not sum to group totals because of missing race/ethnicity data in 1.1% of patients.
cNumbers do not sum to group totals because of missing payment source data in <0.01% of patients.
dNumbers do not sum to group totals because of missing income data in 1.0% of patients.
Over 4.7 (±1.89) years of follow-up, 9701 (1.62%; 95% CI: 1.59%-1.65%) patients were hospitalized for an ischemic stroke. Patients with stroke were older, more likely to be male, and more likely to have a multitude of vascular risk factors (Table 3). Incidence rates of stroke were 4.1 (95% CI: 4.0-4.2) per 1000 person-years after discharge with TNA symptoms, 20.2 (95% CI: 19.3-21.2) per 1000 person-years after TIA, and 1.0 (95% CI: 0.9-1.1) per 1000 person-years after renal colic. By 90 days, the cumulative rate of stroke was 2.9 (95% CI: 2.8-3.1) per 1000 patients with TNA symptoms, 20.8 (95% CI: 18.9-22.8) per 1000 patients with TIA, and 0.3 (95% CI: 0.2-0.4) per 1000 patients with renal colic (Figure 1).
Table 3.
Characteristics of Patients Stratified by Acute Ischemic Stroke.
| Characteristica | Stroke, N = 9701 | No Stroke, N = 588 386 |
|---|---|---|
| Age, mean (SD), years | 70.4 (15.0) | 50.4 (19.0) |
| Female | 5,621 (57.9) | 347,842 (59.1) |
| Raceb | ||
| White | 6,015 (62.8) | 303,133 (52.1) |
| Black | 1,975 (20.6) | 115,795 (19.9) |
| Hispanic | 981 (10.2) | 94,032 (16.2) |
| Asian | 141 (1.5) | 17,129 (2.9) |
| Other | 471 (4.9) | 51,986 (8.9) |
| Payment sourcec | ||
| Medicare | 5,310 (54.7) | 123,185 (20.9) |
| Medicaid | 766 (7.9) | 76,473 (13.0) |
| Private | 3,084 (31.8) | 294,701 (50.1) |
| Self-pay | 472 (4.9) | 82,351 (14.0) |
| Other | 69 (0.7) | 11,671 (2.0) |
| Income quartiled | ||
| 1 | 3,010 (31.4) | 184,617 (31.7) |
| 2 | 2,582 (26.9) | 144,777 (24.9) |
| 3 | 2,167 (22.6) | 132,091 (22.7) |
| 4 | 1,824 (19.0) | 121,038 (20.8) |
| Hypertension | 4,261 (43.9) | 119,689 (20.3) |
| Diabetes | 2,124 (21.9) | 49,385 (8.4) |
| Congestive heart failure | 314 (3.2) | 4,790 (0.8) |
| Coronary heart disease | 1,195 (12.3) | 21,218 (3.6) |
| Peripheral vascular disease | 145 (1.5) | 2,138 (0.4) |
| Chronic obstructive pulmonary disease | 345 (3.6) | 7,869 (1.3) |
| Chronic kidney disease | 319 (3.3) | 4,526 (0.8) |
| Atrial fibrillation | 577 (6.0) | 7,929 (1.4) |
| Tobacco use | 130 (1.3) | 4,099 (0.7) |
| Alcohol use | 359 (3.7) | 22,728 (3.9) |
Abbreviation: SD, standard deviation.
aData are presented as number (%) unless otherwise specified.
bSelf-reported by patients or their surrogates. Numbers do not sum to group totals because of missing race/ethnicity data in 1.1% of patients.
cNumbers do not sum to group totals because of missing payment source data in <0.01% of patients.
dNumbers do not sum to group totals because of missing income quartile data in 1.0% of patients.
Figure 1.
Cumulative acute ischemic stroke rates after transient neurological attack, transient ischemic attack, and renal colic. The cumulative rate of acute ischemic stroke after emergency department discharge for nonspecific neurological symptoms is higher than that after discharge for renal colic but markedly lower than that after discharge for transient ischemic attack, which is also uniquely followed by a short-term increase in stroke rate.
After adjustment for demographics and vascular comorbidities, TNA symptoms were associated with a higher hazard of stroke when compared to renal colic (hazard ratio [HR]: 2.13; 95% CI: 1.90-2.40) but a significantly lower hazard when compared to TIA (HR: 0.47; 95% CI: 0.44-0.50).
In subgroup analysis stratified by age, compared to TIA, TNA was less strongly associated with stroke among patients under 60 years of age (HR: 0.22; 95% CI: 0.19-0.25) compared to those 60 years of age and older (HR: 0.50; 95% CI: 0.47-0.53; P < .001 for interaction). In a sensitivity analysis, altered mental status (HR: 2.81; 95% CI: 2.44-3.24) and nonspecific sensory disturbances (HR: 2.59; 95% CI: 2.27-2.96) appeared slightly more strongly associated with stroke than observed in the primary analysis. An additional sensitivity analysis restricting patients with TNA to those who did not undergo neuroimaging did not change our results. Last, additionally adjusting for a prior hospital-based diagnosis of migraine did not change our results.
Discussion
Patients discharged from the ED with a primary discharge diagnosis code for TNA symptoms experienced an increased long-term rate of ischemic stroke as compared to those discharged with a diagnosis of renal colic, but the absolute rate was low in general and significantly lower than the risk after TIA.
Our findings should be considered in light of several important prior studies. A seminal study on this topic found a slight but nonsignificant association between nonfocal TNA and subsequent ischemic stroke (HR: 1.16; 95 CI: 0.65-2.08).5 However, given the wide CI, this study may have been underpowered (n = 228) to find an association between TNA and ischemic stroke. An association between ED discharge for TNA symptoms and subsequent stroke is plausible given the known association between TNA and incident dementia and vascular death.5 Additionally, over time, certain symptoms previously considered nonspecific, such as isolated vertigo and isolated diplopia, have become recognized as TIA equivalents in terms of the risk of subsequent infarction in the posterior circulation.17-21 This raises the possibility that additional TNA symptoms may have similar clinical and pathophysiological relevance. However, our finding of a low absolute population-level stroke rate after presentation with TNA symptoms challenges recent data, suggesting a high rate of brain infarction in patients with TNA.7,8 Both studies were restricted to patients who were referred to specialized TIA clinics and diagnosed by stroke neurologists, suggesting that the surprisingly high rate of brain infarction may have been due to selection of particularly high-risk patients. Although there may be some high-risk patients within the wider population of those presenting with nonspecific TNA symptoms, such as those with altered mental status, our findings suggest that at a population level and particularly in those under the age of 60, targeting patients with TNA symptoms may not be of substantial yield in reducing stroke risk. Therefore, our findings do not support indiscriminate aggressive risk stratification for patients presenting with nonspecific neurological symptoms, particularly in younger patients.
Our study benefits from its large size, its use of population-level data, the demographically heterogeneous nature of New York State, and the use of several sensitivity analyses to identify subgroups at high risk of stroke among patients presenting with TNA. However, the study has several limitations. First, the study was retrospective in design and relied on administrative claims data. Patients discharged from the ED with a discharge diagnosis of TNA symptoms are likely a heterogeneous group, especially given the low interobserver agreement in making the competing diagnosis of TIA.22 The diagnosis codes for TNA were not validated. This raises the possibility of misclassification errors. For example, patients assigned diagnosis codes for TNA symptoms may in fact have had nonvascular diagnoses such as ophthalmological disorders or neuropathy, which would not be expected to be associated with stroke. As a result, the risk of stroke observed in our cohort of patients with TNA may be an underestimation. Further, there may be intergroup misclassification between TNA and TIA. Substantial intergroup misclassification would be expected to attenuate intergroup differences in stroke risk. Because patients with TNA and TIA were observed to have significantly different stroke risks, we surmise that the likelihood of substantial intergroup misclassification is low. However, any intergroup misclassification error raises the possibility that our findings underestimate the difference in stroke risk between patients with TNA and TIA. In other studies on this subject, TNA and TIA were adjudicated by neurologists; however, they identified patients with TNA from among a group of high-risk patients referred to a TIA clinic,7,8 and adjudicators’ interobserved agreement was low.7 Additionally, discharge diagnosis codes do not reflect the duration of symptoms, acuity of onset, or other clinical features that may be helpful in characterizing a patient’s complaints. Specifically, although previous reports focused on transient neurological symptoms, we cannot comment on the length of time patients had neurologic symptoms in our study. However, we were interested precisely in the overall prognosis of all patients currently found to have symptomatology consistent with TNA upon ED evaluation. The availability of further clinical detail would be unlikely to change our finding that, as a group, patients presenting to and discharged from the ED with nonspecific neurological symptoms face a low risk of future stroke. Second, our data are limited to patients presenting to an ED, so our findings may not be generalizable to other ambulatory settings. Third, patients with acute ischemic stroke who presented with stroke to hospitals outside of New York State were not captured as outcomes. However, this limitation applies equally to all exposure groups. Similarly, our data were not linked to vital statistics data; patients were therefore not censored at the time of out-of-hospital death. The inability to account for the competing risk of out-of-hospital death may result in less robust point estimates of risk in any individual exposure group. However, comparison of stroke risk between groups should not be significantly influenced by this limitation. Last, the rate of stroke after TIA was lower than historical controls,23 likely due to our inclusion of only low-risk patients discharged directly from the ED. This limitation represents a conservative bias because it would likely result in an underestimation of the difference in stroke risk between patients with TIA and TNA symptoms. Alternatively, the rate of stroke after TIA observed in our analysis may be related to improvements in TIA care.24,25
Our findings suggest that patients discharged from the ED with a diagnosis of a nonspecific TNA symptom face a low overall risk of subsequent ischemic stroke. Although further research to risk stratify these patients is warranted, the overall group of patients currently discharged from the ED with TNA symptoms does not appear to represent a high-risk population in need of time-sensitive stroke prevention interventions.
Acknowledgment
The authors are grateful to Monica Chen for administrative assistance.
Footnotes
Declaration of Conflicting Interests: The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Dr Neal S. Parikh is supported by grant T32 NS07153 from the National Institute of Neurological Disorders and Stroke. Dr Hooman Kamel is supported by grants R01NS097443 and K23NS082367 from the National Institute of Neurological Disorders and Stroke.
ORCID iD: Neal S. Parikh, MD 
http://orcid.org/0000-0002-8802-2380
References
- 1. Johnston SC, Gress DR, Browner WS, Sidney S. Short-term prognosis after emergency department diagnosis of TIA. JAMA. 2000;284(22):2901–2906. [DOI] [PubMed] [Google Scholar]
- 2. Hill MD, Yiannakoulias N, Jeerakathil T, Tu JV, Svenson LW, Schopflocher DP. The high risk of stroke immediately after transient ischemic attack: a population-based study. Neurology. 2004;62(11):2015–2020. [DOI] [PubMed] [Google Scholar]
- 3. Kernan WN, Ovbiagele B, Black HR, et al. Guidelines for the prevention of stroke in patients with stroke and transient ischemic attack: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2014;45(7):2160–2236. [DOI] [PubMed] [Google Scholar]
- 4. Lavallée PC, Meseguer E, Abboud H, et al. A transient ischaemic attack clinic with round-the-clock access (SOS-TIA): feasibility and effects. Lancet Neurol. 2007;6(11):953–960. [DOI] [PubMed] [Google Scholar]
- 5. Bos MJ, van Rijn MJ, Witteman JC, Hofman A, Koudstaal PJ, Breteler MM. Incidence and prognosis of transient neurological attacks. JAMA. 2007;298(24):2877–2885. [DOI] [PubMed] [Google Scholar]
- 6. Bots ML, van der Wilk EC, Koudstaal PJ, Hofman A, Grobbee DE. Transient neurological attacks in the general population. Prevalence, risk factors, and clinical relevance. Stroke. 1997;28(4):768–773. [DOI] [PubMed] [Google Scholar]
- 7. van Rooij FG, Vermeer SE, Góraj BM, et al. Diffusion-weighted imaging in transient neurological attacks. Ann Neurol. 2015;78(6):1005–1010. [DOI] [PubMed] [Google Scholar]
- 8. Lavallée PC, Sissani L, Labreuche J, et al. Clinical significance of isolated atypical transient symptoms in a cohort with transient ischemic attack. Stroke. 2017;48(6):1495–1500. [DOI] [PubMed] [Google Scholar]
- 9. Barrett M, Steiner C, Andrews R, Kassed C, Nagamine M. Methodological issues when studying readmissions and revisits using hospital administrative data. HCUP Methods Series Report # 2011-01. https://www.hcup-us.ahrq.gov/reports/methods/2011_01.pdf. Accessed March 9, 2011.
- 10. User guide: HCUP supplemental variables for revisit analysis. Healthcare Cost and Utilization Project (HCUP). December 20, 2017. Agency for Healthcare Research and Quality, Rockville, MD: https://www.hcup-us.ahrq.gov/toolssoftware/revisit/UserGuide_SuppRevisitFilesCD.pdf. Accessed September 20, 2017. [Google Scholar]
- 11. Andrade SE, Harrold LR, Tjia J, et al. A systematic review of validated methods for identifying cerebrovascular accident or transient ischemic attack using administrative data. Pharmacoepidemiol Drug Saf. 2012;21(suppl 1):100–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Atzema CL, Grewal K, Lu H, Kapral MK, Kulkarni G, Austin PC. Outcomes among patients discharged from the emergency department with a diagnosis of peripheral vertigo. Ann Neurol. 2015;79(1):32–41. [DOI] [PubMed] [Google Scholar]
- 13. Tirschwell DL, Longstreth WT., Jr Validating administrative data in stroke research. Stroke. 2002;33(10):2465–2470. [DOI] [PubMed] [Google Scholar]
- 14. Johnston SC, Rothwell PM, Nguyen-Huynh MN, et al. Validation and refinement of scores to predict very early stroke risk after transient ischaemic attack. Lancet. 2007;369(9558):283–292. [DOI] [PubMed] [Google Scholar]
- 15. Osler M, Wium-Andersen IK, Jørgensen MB, Jørgensen TSH, Wium-Andersen MK. Migraine and risk of stroke and acute coronary syndrome in two case-control studies in the Danish population. Clin Epidemiol. 2017;9:439–449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Friedman BW, West J, Vinson DR, Minen MT, Restivo A, Gallagher EJ. Current management of migraine in US emergency departments: an analysis of the National Hospital Ambulatory Medical Care Survey. Cephalalgia. 2015;35(4):301–309. [DOI] [PubMed] [Google Scholar]
- 17. Lee CC, Su YC, Ho HC, et al. Risk of stroke in patients hospitalized for isolated vertigo: a four-year follow-up study. Stroke. 2011;42(1):48–52. [DOI] [PubMed] [Google Scholar]
- 18. Lee CC, Ho HC, Su YC, et al. Increased risk of vascular events in emergency room patients discharged home with diagnosis of dizziness or vertigo: a 3-year follow-up study. PLoS One. 2012;7(4):e35923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Hoshino T, Nagao T, Mizuno S, Shimizu S, Uchiyama S. Transient neurological attack before vertebrobasilar stroke. J Neurol Sci. 2013;325(1-2):39–42. [DOI] [PubMed] [Google Scholar]
- 20. Paul NL, Simoni M, Rothwell PM. Transient isolated brainstem symptoms preceding posterior circulation stroke: a population-based study. Lancet Neurol. 2013;12(1):65–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Kim AS, Fullerton HJ, Johnston SC. Risk of vascular events in emergency department patients discharged home with diagnosis of dizziness or vertigo. Ann Emerg Med. 2011;57(1):34–41. [DOI] [PubMed] [Google Scholar]
- 22. Castle J, Mlynash M, Lee K, et al. Agreement regarding diagnosis of transient ischemic attack fairly low among stroke-trained neurologists. Stroke. 2010;41(7):1367–1370. [DOI] [PubMed] [Google Scholar]
- 23. Wu CM, McLaughlin K, Lorenzetti DL, Hill MD, Manns BJ, Ghali WA. Early risk of stroke after transient ischemic attack: a systematic review and meta-analysis. Arch Intern Med. 2007;167(22):2417–2422. [DOI] [PubMed] [Google Scholar]
- 24. Stead LG, Suravaram S, Bellolio MF, et al. An assessment of the incremental value of the ABCD2 score in the emergency department evaluation of transient ischemic attack. Ann Emerg Med. 2011;57(1):46–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Rothwell PM, Giles MF, Chandratheva A, et al. Effect of urgent treatment of transient ischaemic attack and minor stroke on early recurrent stroke (EXPRESS study): a prospective population-based sequential comparison. Lancet. 2007;370(9596):1432–1442. [DOI] [PubMed] [Google Scholar]