de Lusignan Simon, McGee Christopher, Webb Rebecca, Joy Mark, Byford Rachel, Yonova Ivelina, Hriskova Mariya, Ferreira Filipa, Elliot Alex J, Smith Gillian, Rafi Imran (2018) Conurbation, Urban, and Rural Living as Determinants of Allergies and Infectious Diseases: Royal College of General Practitioners Research and Surveillance Centre Annual Report 2016-2017,JMIR Public Health and Surveillance4(4)e11354
Background: Living in a conurbation, urban, or rural environment is an important determinant of health. For example, conurbation and rural living is associated with increased respiratory and allergic conditions, whereas a farm or rural upbringing has been shown to be a protective factor against this.
Objective: The objective of the study was to assess differences in general practice presentations of allergic and infectious disease in those exposed to conurbation or urban living compared with rural environments.
Methods: The population was a nationally representative sample of 175 English general practices covering a population of over 1.6 million patients registered with sentinel network general practices. General practice presentation rates per 100,000 population were reported for allergic rhinitis, asthma, and infectious conditions grouped into upper and lower respiratory tract infections, urinary tract infection, and acute gastroenteritis by the UK Office for National Statistics urban-rural category. We used multivariate logistic regression adjusting for age, sex, ethnicity, deprivation, comorbidities, and smoking status, reporting odds ratios (ORs) with 95% CIs.
Results: For allergic rhinitis, the OR was 1.13 (95% CI 1.04-1.23; P=.003) for urban and 1.29 (95% CI 1.19-1.41; P<.001) for conurbation compared with rural dwellers. Conurbation living was associated with a lower OR for both asthma (OR 0.70, 95% CI 0.67-0.73; P<.001) and lower respiratory tract infections (OR 0.94, 95% CI 0.90-0.98; P=.005). Compared with rural dwellers, the OR for upper respiratory tract infection was greater in urban (OR 1.06, 95% CI 1.03-1.08; P<.001) but no different in conurbation dwellers (OR 1.00, 95% CI 0.97-1.03; P=.93). Acute gastroenteritis followed the same pattern: the OR was 1.13 (95% CI 1.01-1.25; P=.03) for urban dwellers and 1.04 (95% CI 0.93-1.17; P=.46) for conurbation dwellers. The OR for urinary tract infection was lower for urban dwellers (OR 0.94, 95% CI 0.89-0.99; P=.02) but higher in conurbation dwellers (OR 1.06, 95% CI 1.00-1.13; P=.04).
Conclusions: Those living in conurbations or urban areas were more likely to consult a general practice for allergic rhinitis and upper respiratory tract infection. Both conurbation and rural living were associated with an increased risk of urinary tract infection. Living in rural areas was associated with an increased risk of asthma and lower respiratory tract infections. The data suggest that living environment may affect rates of consultations for certain conditions. Longitudinal analyses of these data would be useful in providing insights into important determinants.
The use of deep learning is becoming increasingly important in the analysis of medical data such as pattern recognition for classification. The use of primary healthcare computational medical records (CMR) data is vital in prediction of infection prevalence across a population, and decision making at a national scale. To date, the application of machine learning algorithms to CMR data remains under-utilized despite the potential impact for use in diagnostics or prevention of epidemics such as outbreaks of influenza. A particular challenge in epidemiology is how to differentiate incident cases from those that are follow-ups for the same condition. Furthermore, the CMR data are typically
heterogeneous, noisy, high dimensional and incomplete, making automated analysis difficult. We introduce a methodology for converting heterogeneous data such that it is compatible with a deep autoencoder for reduction of CMR data. This approach provides a tool for real time visualization of these high dimensional data, revealing previously unknown dependencies and clusters. Our unsupervised nonlinear reduction method can be used to identify the features driving the formation of these clusters that can aid decision making in healthcare applications. The results in this work demonstrate that our methods can cluster more than 97.84% of the data (clusters >5 points) each of which is uniquely described by three attributes in the data: Clinical System (CMR system), Read Code (as recorded) and Read Term (standardized coding). Further, we propose the use of Shannon Entropy as a means to analyse the dispersion of clusters and the contribution from the underlying attributes to gain further insight from the data. Our results demonstrate that Shannon Entropy is a useful metric for analysing
both the low dimensional clusters of CMR data, and also the features in the original heterogeneous data. Finally, we find that the entropy of the low dimensional clusters are directly representative of the entropy of the input data (Pearson Correlation = 0.99, R2 = 0.98) and therefore the reduced data from the deep autoencoder is reflective of the original CMR data variability.
The finding that drugs and metabolites can be detected from fingerprints is of potential relevance to forensic science and as well as toxicology and clinical testing. However, discriminating between dermal contact and ingestion of drugs has never been verified experimentally. The inability to interpret the result of finding a drug or metabolite in a fingerprint has prevented widespread adoption of fingerprints in drug testing and limits the probative value of detecting drugs in fingermarks. A commonly held belief is that the detection of metabolites of drugs of abuse in fingerprints can be used to confirm a drug has been ingested. However, we show here that cocaine and its primary metabolite, benzoylecgonine, can be detected in fingerprints of non-drug users after contact with cocaine. Additionally, cocaine was found to persist above environmental levels for up to 48 hours after contact. Therefore the detection of cocaine and benzoylecgonine (BZE) in fingermarks can be forensically significant, but do not demonstrate that a person has ingested the substance. In contrast, the data here shows that a drug test from a fingerprint (where hands can be washed prior to donating a sample) CAN distinguish between contact and ingestion of cocaine. If hands were washed prior to giving a fingerprint, BZE was detected only after the administration of cocaine. Therefore BZE can be used to distinguish cocaine contact from cocaine ingestion, provided donors wash their hands prior to sampling. A test based on the detection of BZE in at least one of two donated fingerprint samples has accuracy 95%, sensitivity 90% and specificity of 100% (n = 86).