Information extraction from clinical and radiology notes for liver cancer staging

Thesis (Ph.D.)--University of Washington, 2016-06

Approaches and Strategy for Cancer Research and Surveillance Data: Integration, Information Pipeline, Data Models, and Informatics Opportunities

Thesis (Ph.D.)--University of Washington, 2016-04

Proton pump inhibitors and the risk of fractures in older adults: a population-based retrospective cohort study

Thesis (Master's)--University of Washington, 2016-06

Self-reported difficulty in conceiving as a measure of infertility

BACKGROUND: This study aimed to explore the meaning and potential use of women's self-reported difficulties in conceiving as a measure of infertility in epidemiological studies, and to compare women's stated reasons for infertility with information in their medical records. METHODS: Data were available from a population-based case-control study of ovarian cancer involving 1638 women. The sensitivity and specificity of women's self-reported infertility were calculated against their estimated fertility status based on detailed reproductive histories. Self-reported reasons for infertility were compared with diagnoses documented in women's medical records. RESULTS: The sensitivity of women's self-reported difficulty in conceiving was 66 and 69% respectively when compared with calendar-derived and self-reported times taken trying to conceive; its specificity was 95%. Forty-one (23%) of the 179 women for whom medical records were available had their self-reported fertility problem confirmed. Self-reported infertility causes could be compared with diagnoses in medical records for only 22 of these women. CONCLUSIONS: Self-reported difficulty conceiving is a useful measure of infertility for quantifying the burden of fertility problems experienced in the community. Validation of reasons for infertility is unlikely to be feasible through examination of medical records. Improved education of the public regarding the availability and success rates of infertility treatments is proposed.

Effects of patient-focused care on seclusion in a psychiatric intensive care unit

The objective of the study was to evaluate whether the introduction of patient-focused nursing care affected the number of seclusions and the length of time patients spent in seclusion, in an acute psychiatric unit. The study used a pre-intervention–post-intervention design and was conducted in an eight-bed locked unit within a large regional general hospital in Queensland, Australia. The medical records of all people who were secluded as part of their management while in hospital, during two 6-month periods, were retrospectively reviewed. Changes to the ways in which nurses conducted their daily activities were implemented during the time between the data collection periods. There were no differences between the groups with respect to the number of times a patient was secluded. However, following implementation of patient-focused care, there was a reduction in the length of time for which patients were secluded. The only change in medication administration was that post-implementation, Haloperidol was used in fewer seclusion episodes. The findings indicate that changes to nursing practice may result in closer monitoring of patients and a reduction in the time patients spend secluded in acute inpatient psychiatric settings.

Validity of self-reported skin screening histories

Screening by whole-body clinical skin examination may improve early diagnosis of melanoma and reduce mortality, but objective scientific evidence of this is lacking. As part of a randomized controlled trial of population screening for melanoma in Queensland, Australia, the authors assessed the validity of self-reported history of whole-body skin examination and factors associated with accuracy of recall among 2,704 participants in 2001. Approximately half of the participants were known to have undergone whole-body skin examination within the past 3 years at skin screening clinics conducted as part of the randomized trial. All positive and negative self-reports were compared with screening clinic records. Where possible, reports of skin examinations conducted outside the clinics were compared with private medical records. The validity of self-reports of whole-body skin examination in the past 3 years was high: Concordance between self-reports and medical records was 93.7%, sensitivity was 92.0%, and specificity was 96.3%. Concordance was lower (74.3%) for self-reports of examinations conducted in the past 12 months, and there was evidence of telescoping in recall for this more recent time frame. In multivariate analysis, women and younger participants more accurately recalled their history of skin examinations. Participants with a history of melanoma did not differ from other participants in their accuracy of recall.

A cost-effectiveness analysis of two rehabilitation support services for women with breast cancer

The purpose of this research was to estimate the cost-effectiveness of two rehabilitation interventions for breast cancer survivors, each compared to a population-based, non-intervention group (n = 208). The two services included an early home-based physiotherapy intervention (DAART, n = 36) and a group-based exercise and psychosocial intervention (STRETCH, n = 31). A societal perspective was taken and costs were included as those incurred by the health care system, the survivors and community. Health outcomes included: (a) 'rehabilitated cases' based on changes in health-related quality of life between 6 and 12 months post-diagnosis, using the Functional Assessment of Cancer Therapy - Breast Cancer plus Arm Morbidity (FACT-B+4) questionnaire, and (b) quality-adjusted life years (QALYs) using utility scores from the Subjective Health Estimation (SHE) scale. Data were collected using self-reported questionnaires, medical records and program budgets. A Monte-Carlo modelling approach was used to test for uncertainty in cost and outcome estimates. The proportion of rehabilitated cases was similar across the three groups. From a societal perspective compared with the non-intervention group, the DAART intervention appeared to be the most efficient option with an incremental cost of $1344 per QALY gained, whereas the incremental cost per QALY gained from the STRETCH program was $14,478. Both DAART and STRETCH are low-cost, low-technological health promoting programs representing excellent public health investments.

The relationship between acute alcohol consumption and consequent injury type

Aims: The aim of this study was to quantify the relationship between acute alcohol consumption and injury type (nature of injury, body region injured), while adjusting for the effect of known confounders (i.e. demographic and situational variables, usual drinking patterns, substance use and risk-taking behaviour). Methods: A cross-sectional study was conducted between October, 2000 and October, 2001 of patients aged >= 15 years presenting to a Queensland Emergency Department for treatment of an injury sustained in the preceding 24 h. There were three measures of acute alcohol consumption: drinking setting, quantity, and beverage type consumed in the 6 h prior to injury. Two variables were used to quantify injury type: nature of injury (fracture/dislocation, superficial, internal, and CNS injury) and body part injured (head/neck, facial, chest, abdominal, external, and extremities). Both were derived from patient medical records. Results: Five hundred and ninety three patients were interviewed. Logistic regression analyses indicated that, after controlling for relevant confounding variables, there was no significant association between any of the three measures of acute alcohol consumption and injury type. Conclusions: The effects of acute alcohol consumption are not specific to injury type. Interventions aimed at reducing the incidence of alcohol-related injury should not be targeted at specific injury types.

Assessing the effectiveness of a mammography screening service

Background: Trials have shown that mammography screening reduces mortality and probably decreases morbidity related to breast cancer. Methods: We assessed whether the major mammography service in Western Australia (BreastScreen WA) is likely to reduce mortality by comparing prognostic variables between screen-detected and other cases of breast cancer diagnosed in 1999. We assessed likely reductions in morbidity by comparing treatments received by these two groups. To confirm mortality and morbidity reduction, we also compared prognostic variables and treatments with targets. Information on demographic variables, tumour characteristics at presentation and treatments were collected from medical records for all incident cases of breast cancer in Western Australia in 1999. We matched cases with the Western Australian Cancer Registry records to determine which cases had been detected by BreastScreen WA. Results: BreastScreen WA achieved the targets for mortality reduction. Tumours detected by BreastScreen WA were smaller in size, less likely to have vascular invasion, of lower histological grade and were more likely to be ductal carcinoma in situ alone without invasive carcinoma. Oestrogen receptor status was more likely to be positive, the difference in progesterone status was not significant, and lymph node involvement tended to be lower. BreastScreen WA patients were treated more often with local therapy and less often with systemic therapy, and the proportion of patients treated with breast-conserving surgery was close to the target for minimizing morbidity in breast cancer. Conclusion: Mammographic detection of breast cancer by BreastScreen WA is associated with reduced breast cancer morbidity and a more favourable prognosis.

Validity of verbal autopsy procedures for determining cause of death in Tanzania

Objectives: To validate verbal autopsy (VA) procedures for use in sample vital registration. Verbal autopsy is an important method for deriving cause-specific mortality estimates where disease burdens are greatest and routine cause-specific mortality data do not exist. Methods: Verbal autopsies and medical records (MR) were collected for 3123 deaths in the perinatal/neonatal period, post-neonatal < 5 age group, and for ages of 5 years and over in Tanzania. Causes of death were assigned by physician panels using the International Classification of Disease, revision 10. Validity was measured by: cause-specific mortality fractions (CSMF); sensitivity; specificity and positive predictive value. Medical record diagnoses were scored for degree of uncertainty, and sensitivity and specificity adjusted. Criteria for evaluating VA performance in generating true proportional mortality were applied. Results: Verbal autopsy produced accurate CSMFs for nine causes in different age groups: birth asphyxia; intrauterine complications; pneumonia; HIV/AIDS; malaria (adults); tuberculosis; cerebrovascular diseases; injuries and direct maternal causes. Results for 20 other causes approached the threshold for good performance. Conclusions: Verbal autopsy reliably estimated CSMFs for diseases of public health importance in all age groups. Further validation is needed to assess reasons for lack of positive results for some conditions.

Effects of care pathways on stroke care practices at regional hospitals

Background: Our previous work identified deficiencies in stroke care practices at regional hospitals in comparison to standards suggested by published stroke care guidelines. These deficiencies might be improved by the implementation of clinical pathways. The aim of this study was to assess changes in acute stroke care practices following the implementation of stroke care pathways at four regional Queensland hospitals. Methods: The medical records of two cohorts of 120 patients with a discharge diagnosis of stroke or transient ischaemic attack were retrospectively audited before and after implementation of stroke care pathways to identify differences in the use of acute interventions, investigations and secondary prevention strategies. Results: Following pathway implementation there were clinically important, but not statistically significant, increases in the rates of swallow assessment, allied health assessment (significant for occupational therapy, P = 0.04) and use of deep vein thrombosis prevention strategies (also significant, P = 0.006). Fewer patients were discharged on no anti-thrombotic therapy (statistically significant in the subgroup of patients with atrial fibrillation, P = 0.02). Only 37% of the patients audited were actually enrolled on the pathway. Among this subgroup there were significant increases in the rates of swallow assessment (first 24 h, P = 0.01; any time during admission, P = 0.0001), allied health assessments (all P < 0.05), estimation of blood glucose level (P = 0.0015) and the use of deep vein thrombosis prevention strategies (P = 0.0003). Conclusion: Stroke care pathways appear to improve the process of care. Whether this influences outcomes such as mortality, functional and neurological recovery, the incidence of complications, length of stay or the cost of care was beyond the scope of this study and will require further examination.

Beta-lactam allergy in adults with cystic fibrosis

Background: Allergic reactions to one or more beta-lactam antibiotic can pose a management problem in patients with cystic fibrosis (CF), and may limit antibiotic choice. Method: The aim of this study was to assess the prevalence of allergy to anti-pseudomonal beta-lactam antibiotics in an adult CF centre and to assess variables, which may contribute to the development of allergic reactions. A questionnaire-based interview and a review of medical records were performed. Results: Of the 150 patients, 54 (36%) had allergic reactions to one or more beta-lactam antibiotics and 20 (19%) had allergic reactions to multiple beta-lactam antibiotics. The proportion of patients allergic to specific beta-lactam antibiotics varied from 10% to 26%. Rates of reactions were highest for penicillins and cephalosporins, intermediate for carbepenems and lowest for aztreonam. Of all reactions, 40% occurred within 24 h of the commencement of an individual antibiotic course. Patients with one or more beta-lactam allergic reactions had received greater cumulative exposure (p < 0.0001), were older (p=0.016) and had lower lung function (p=0.037) than patients without a history of beta-lactam allergy. Cystic Fibrosis transmembrane regulator (CFTR) status, gender, peripheral blood eosinophil count and total IgE concentrations were not different in patients with allergic reactions. Conclusions: This study demonstrates that the prevalence of allergic reactions to beta-lactam antibiotics is high in adults with CF. Increasing age; cumulative exposure and decreasing FEV1 were associated with the development of allergy. (C) 2006 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved.

Use of routine data for determination of the population pharmacokinetics and enteral biovailability of phenytoin in neonates and infants with seizures

Objective: To investigate the population pharmacokinetics and the enteral bioavailability of phenytoin in neonates and infants with seizures. Methods: Data (5 mg kg-1 day-1) from 83 patients were obtained retrospectively from the medical records following written ethical approval. A one-compartment model was fitted to the data using NONMEM with FOCE-interaction. Between-subject variability (BSV) and interoccasion variability (IOV) were modelled exponentially together with a log transform-both-sides exponential residual unexplained variance (RUV) model. Covariates in nested models were screened for significance (X2, 1, 0.01). Model validity was determined by bootstrapping with replacement (N=500 samples) from the dataset. Results: The parameters of final pharmacokinetic were: Clearance (L h-1) = 0.826.(current Weight [kg]/70)0.75.(1+0.0692.(Postnatal age [days]-11)); Volume of distribution (L) = 74.2.(current Weight [kg]/70); Enteral bioavailability = 0.76; Absorption rate constant (h-1) = 0.167. BSV for clearance and volume of distribution were 74.2% and 65.6%, respectively. The IOV in clearance was 54.4%. The RUV was 51.1%. Final model parameters deviated from mean bootstrap estimates by

ANOMALIAS DENTAIS ASSOCIADAS ÀS AGENESIAS

O objetivo deste estudo retrospectivo foi avaliar a prevalência de anomalias de posição (irrupção ectópica de canino superior para palatino, transposição dental, distoangulação de segundos pré-molares inferiores, mesioangulação de segundo molar inferior permanente e infraoclusão de molares decíduos), de número (supranumerários) e de tamanho (microdontias) em pacientes com agenesias de dentes permanentes, comparando-as com as prevalências esperadas para a população em geral, além de testar a hipótese de que pacientes com agenesia de dentes permanentes apresentem uma prevalência aumentada de hipoplasia de esmalte. Para tanto, a amostra deste estudo foi composta por 351 pacientes, com a presença de agenesia de no mínimo um dente permanente, na faixa etária entre 8 e 30 anos e com prontuários clínicos preenchidos. A amostra foi coletada a partir do exame das documentações ortodônticas pertencentes aos arquivos de uma escola de aperfeiçoamento profissional em Ortodontia, de uma clínica radiológica odontológica e de consultórios particulares de ortodontistas. O material de estudo englobou radiografias panorâmicas e periapicais, modelos de gesso, fotografias intra e extraorais e prontuários clínicos devidamente preenchidos. Inicialmente foi analisada a reprodutibilidade das avaliações pela porcentagem de concordância utilizando Kappa, com intervalo de confiança de 95%. O teste de qui-quadrado foi utilizado para comparar as prevalências de agenesias e anomalias na amostra com as prevalências esperadas segundo a literatura científica, considerando o nível de significância de 5%. Analisou-se, ainda, o grau das associações pela razão de chances ( odds ratio ) e o respectivo intervalo de confiança de 95%. A prevalência de agenesias dentais na amostra, excluindo os terceiros molares, foi de 88,6%. Dos 351 pacientes, 128 (36,4%) apresentavam agenesia no arco maxilar, 108 (30,8%) no mandibular e 115 (32,8%) nos dois arcos. Em relação ao hemiarco maxilar esquerdo, 52,4% apresentavam agenesia, no direito 55,0%, no mandibular esquerdo 48,7% e no direito 47,3%. Das anomalias associadas avaliadas, 28,5% microdontia, 28,2% hipoplasia de esmalte, 7,4% apresentavam irrupção ectópica de canino superior por palatino, 6,6% distoangulação, 3,9% transposição de canino/pré-molar superior, 4,3% infraoclusão, 3,7 supranumerário, 3,7% mesioangulação, 0,6% transposição de incisivo/canino inferior, e, quando comparadas com a população em geral, observou que 96,1 vezes mais chance de apresentar mesioangulação do segundo molar inferior; 34,6 vezes mais chance de apresentar distoangulação; 15,9 vezes mais chance de apresentar transposição canino/pré-molar superior; 14,3 vezes mais chance de apresentar transposição de incisivo/canino inferior; 9 vezes mais chance de hipoplasia; a microdontia do incisivo lateral apresentou 8,1 vezes mais chance; 5,2 vezes mais chance de apresentar irrupção ectópica do canino superior por palatino, e, em relação à infraoclusão, apresentando uma menor chance do que a população geral. A partir dos resultados obtidos, verificou-se uma forte associação entre a agenesia de dentes permanentes, correlacionando com outras anomalias dentais importantes. Foi constatado de que pacientes com agenesia de dentes permanentes apresentam uma prevalência aumentada de hipoplasia de esmalte e de que agenesias e outras anomalias associadas apresentam-se interligadas geneticamente entre si.

COMPARAÇÃO DE MEDIDAS CEFALOMÉTRICAS DOS ESPAÇOS AÉREOS NASO E OROFARÍNGEO ENTRE RESPIRADORES NASAIS E ORAIS

A literatura tem evidenciado que o modo respiratório alterado influencia nos aspectos estruturais e funcionais do sistema estomatognático. Objetivo: Comparar as medidas cefalométricas dos espaços aéreos naso e orofaríngeo entre respiradores nasais e com modo respiratório alterado (orais ou oronasais). Método: Este estudo de caráter exploratório e retrospectivo utilizou os prontuários de pacientes atendidos na Clínica de Ortodontia, compreendendo o período de 2000 a 2009, sendo que dos 87 indivíduos (má oclusão Classe I de Angle) que compuseram a amostra, 55 eram respiradores nasais (RN) e 32 apresentavam modo respiratório alterado (RON - oronasal ou oral), 47 pertenciam ao sexo feminino e 40 ao masculino, com idades entre 8,75 e 22 anos (média de 14,78 anos). Foram analisadas as telerradiografias em norma lateral, não submetidos a tratamento ortodôntico ou fonoaudiológica prévios. As variáveis de análise foram: idade, sexo, raça, modo respiratório e análise dos espaços naso e orofaringeano pelo método adotado por Zanelato. Por meio de análise estatística (teste t de Student e quiquadrado), os resultados foram interpretados quanto à sua significância. Resultados: Houve diferença estatisticamente significante quanto à idade dos grupos (RN>RON); de prevalência de homens (maior no grupo RON), de leucodermas em ambos os grupos, de via aérea súpero-anterior (menor no grupo RON) e tonsila faríngea (maior no grupo RON); e de tamanho de tonsila faríngea em melanodermas. As demais variáveis de análise não evidenciaram diferenças estatisticamente significantes. Conclusão: As medidas cefalométricas do espaço nasofaringeano (menor no grupo RON) e da tonsila faríngea (maior no grupo RON) diferiram entre os grupos e o espaço orofaringeano não se mostrou diferente entre os grupos. Demais fatores, que não exclusivamente o espaço nasofaríngeo e a tonsila faríngea, podem predispor o indivíduo à respiração oronasal ou predominantemente oral e o ortodontista, em sua rotina clínica, pode avaliar objetivamente as particularidades de cada caso e, desta forma, proceder com condutas adequadas o mais precocemente possível.