Ability of verbal autopsy data to detect deaths due to uncontrolled hyperglycaemia: testing existing methods and development and validation of a novel weighted score

Blackstock, Sarah; Witham, Miles D.; Wade, Alisha N.; Crampin, Amelia C.; Beran, David; Davies, Justine

doi:10.1136/bmjopen-2018-026331

Cited by 4 publications

(5 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many TCCs could be prevented through the reduction of risk factors and associated diseases, such as adequate treatment of diabetes mellitus reducing the TCC of diabetic ketoacidosis. 51 Additionally, primary care providers should be able to recognise TCC and appropriately refer. The focus of VA on events proximate to death does not allow capture of data on primary prevention efforts that could have prevented TCC; further studies of health service availability in the region are needed to provide data regarding this.…”

Section: Bmj Global Healthmentioning

confidence: 99%

Time-critical conditions: assessment of burden and access to care using verbal autopsy in Agincourt, South Africa

et al. 2020

Self Cite

View full text Add to dashboard Cite

BackgroundTime-critical conditions (TCC) are estimated to cause substantial mortality in low and middle-income countries. However, quantification of deaths and identification of contributing factors to those deaths are challenging in settings with poor health records.AimTo use verbal autopsy (VA) data from the Agincourt health and sociodemographic surveillance system in rural South Africa to quantify the burden of deaths from TCC and to evaluate the barriers in seeking, reaching and receiving quality care for TCC leading to death.MethodologyDeaths from 1993 to 2015 were analysed to identify causality from TCC. Deaths due to TCC were categorised as communicable, non-communicable, maternal, neonatal or injury-related. Proportion of deaths from TCC by age, sex, condition type and temporal trends was described. Deaths due to TCC from 2012 to 2015 were further examined by circumstances of mortality (CoM) indicators embedded in VA. Healthcare access, at illness onset and during the final day of life, as well as place of death, was extracted from free text summaries. Summaries were also analysed qualitatively using a Three Delays framework to identify barriers to healthcare.ResultsOf 15 305 deaths, 5885 (38.45%) were due to TCC. Non-communicable diseases were the most prevalent cause of death from TCC (2961/5885 cases, 50.31%). CoM indicators highlighted delays in a quarter of deaths due to TCC, most frequently in seeking care. The most common pattern of healthcare access was to die outwith a facility, having sought no healthcare (409/1324 cases, 30.89%). Issues in receipt of quality care were identified by qualitative analysis.ConclusionTCCs are responsible for a substantial burden of deaths in this rural South African population. Delays in seeking and receiving quality care were more prominent than those in reaching care, and thus further research and solution development should focus on healthcare-seeking behaviour and quality care provision.

show abstract

Section: Bmj Global Healthmentioning

confidence: 99%

Time-critical conditions: assessment of burden and access to care using verbal autopsy in Agincourt, South Africa

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…The passage of time between the death and the data collection could exacerbate recall errors. Furthermore, in past studies, the InterVA4 software, under predicted diabetes as the cause of death as compared to diagnostic symptom classi cation [18].…”

Section: Potential Sources Of Biasmentioning

confidence: 99%

Examination of Mortality Due to Diabetes and Assessment of Diabetes-Related Healthcare Services in Rural Western Kenya

Purdy

Sifuna

Otieno

et al. 2020

Preprint

View full text Add to dashboard Cite

BackgroundThe prevalence of diabetes is increasing in low-and middle- income countries, due to the adoption of a western diet and decreased physical activity. Diabetes is often underdiagnosed, and the management of the disease is resource intensive. In this paper, we examine the burden of diabetes on a rural population in western, Kenya and assess the ability of the health care system to diagnose, manage, and treat diabetes. MethodsWe utilized verbal autopsies from the Kombewa Health and Demographic Surveillance Site in rural western Kenya from 2011-2018 to measure the burden of diabetes among the deceased. We classified deaths as either primarily caused by diabetes according to verbal autopsy or as a death of any cause that reported having a previous diagnosis of diabetes. We also conducted a survey of health facilities to measure the capacity of the health system to prevent, diagnose, and manage diabetes.ResultsFrom 2011-2018, 85 people died with diabetes as the verbal autopsy reported primary cause in the region (1.8% of verbal autopsies). An additional 4.6% of verbal autopsies indicated the deceased had a previous known diagnosis of diabetes. Deaths with diabetes as the primary cause of death increased with age and were more likely among men than women. Of the 23 surveyed health facilities, 26% regularly screened for diabetes and 39% managed diabetes in their patients. We found a lack of screening and resources to consistently treat patients with diabetes. All facilities reported not having the full range of medications indicated by the Kenya Essential Medication List to treat diabetes. ConclusionsOur results suggest that the undetected burden of diabetes may be greater in rural Western Kenya than previous country-wide studies suggested. Additionally, our results demonstrate that most health facilities in the region do not have the capacity to screen for diabetes nor do they have stocks of the medications to treat diabetes. This lack of care results in patients being referred to larger hospitals for treatment. As the prevalence of diabetes increases in Kenya, and other low- and middle-income countries, improved detection and treatment of diabetes will be important to limiting the deleterious chronic damage caused by undiagnosed and uncontrolled diabetes.

show abstract

“…This lack of diagnosis is compounded by the absence of statistics on diabetes-related deaths in many lower-and-middle-income countries (LMIC). The mortality data that is available is frequently derived from hospital records, and it possibly understates mortality from fatalities that occurring outside of hospital [2]. People with diabetes may develop chronic complications such as neuropathy, nephropathy and retinopathy, uncontrolled hyperglycaemia, and can also develop acute potentially fatal complications such as diabetic ketoacidosis (DKA) and hyperosmolar hyperglycaemic syndrome (HHS) [3,4].…”

Section: Introductionmentioning

confidence: 99%

“…The machine learning techniques of logistic regression, random forest, XGBoost and neural networks were employed to automate COD classification in the three feature settings. We relied on data labeled by physician experts [2] as our gold standard for determining whether a death had been caused by uncontrolled hyperglycaemia (positive; coded as 1) or not (negative; coded as 0).…”

Section: Introductionmentioning

confidence: 99%

Using Machine Learning to Fuse Verbal Autopsy Narratives and Binary Features in the Analysis of Deaths from Hyperglycaemia

Manaka¹,

Zyl²,

Wade³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Lower-and-middle income countries are faced with challenges arising from a lack of data on cause of death (COD), which can limit decisions on population health and disease management. A verbal autopsy (VA) can provide information about a COD in areas without robust death registration systems. A VA consists of structured data, combining numeric and binary features, and unstructured data as part of an openended narrative text. This study assesses the performance of various machine learning approaches when analyzing both the structured and unstructured components of the VA report. The algorithms were trained and tested via cross-validation in the three settings of binary features, text features and a combination of binary and text features derived from VA reports from rural South Africa. The results obtained indicate narrative text features contain valuable information for determining COD and that a combination of binary and text features improves the automated COD classification task.

show abstract

Ability of verbal autopsy data to detect deaths due to uncontrolled hyperglycaemia: testing existing methods and development and validation of a novel weighted score

Cited by 4 publications

References 33 publications

Time-critical conditions: assessment of burden and access to care using verbal autopsy in Agincourt, South Africa

Time-critical conditions: assessment of burden and access to care using verbal autopsy in Agincourt, South Africa

Examination of Mortality Due to Diabetes and Assessment of Diabetes-Related Healthcare Services in Rural Western Kenya

Using Machine Learning to Fuse Verbal Autopsy Narratives and Binary Features in the Analysis of Deaths from Hyperglycaemia

Contact Info

Product

Resources

About