Patterns and trends of medicinal poisoning substances: a population-based cohort study of injuries in 0-11 year old children from 1998-2018

BACKGROUND

There have been sharp increases in antidepressant and opioid prescriptions over the last 10 years, as well as increased over-the-counter medicine availability. However, the impact on childhood medicinal poisonings rates, particularly by socioeconomic deprivation is unclear. This study reports population level medicinal poisoning substance patterns in England among children aged 0-11 years, helping inform safety advice and poisoning prevention interventions.

METHODS

An open cohort study of 1,489,620 0-11 year olds was conducted from 1998 to 2018, using the Clinical Practice Research Datalink, to examine inpatient hospital admissions for poisoning. Incidence rates and adjusted incidence rate ratios (aIRR) were calculated for poisoning substance groups by age, sex, socio-economic deprivation and year.

RESULTS

3,685 medicinal poisoning hospital admissions were identified. The most common substances were paracetamol (33.2%), dependence/withdrawal risk drugs (DWRD - antidepressants, opioids, gabapentinoids, benzodiazepines) (13.5%) and other over-the-counter (OTC) analgesics/anti-common cold drugs (13.0%). Over the study period DWRD poisonings decreased 33% (aIRR 0.67, 95%CI 0.50-0.90 comparing 2013/14-2017/18 to 1998/99-2002/03), while paracetamol poisonings increased 43% (aIRR 1.43, 95%CI 1.20-1.70 for the same periods), with no change in incidence rates for other OTC drugs (aIRR 0.82, 95% CI 0.60-1.12) or all medications combined (aIRR 0.97, 95%CI 0.88-1.07). A gradient in poisonings by area-level socioeconomic deprivation was shown for all medications (aIRR 1.32, 95%CI 1.18-1.47 for most deprived compared to least deprived quintile), and DWRDs (aIRR 2.03, 95%CI 1.42-2.88 for 4th most deprived quintile and aIRR 1.88, 95%CI 1.32-2.66 for 5th most deprived quintile, compared to least deprived quintile), but not for paracetamol or other OTC drug poisonings.

CONCLUSIONS

Poisonings from DWRDs decreased by 33%, while paracetamol poisonings increased by 43% during the study period. There was a gradient by area-level socioeconomic deprivation in prescribed medication poisonings, including drugs with withdrawal/dependence risk, but not OTC medication poisonings. Households in more socioeconomically deprived areas have the potential to benefit most from measures to improve safe storage of medicines and are likely to require targeted interventions providing education and safety equipment. In addition, universal promotion of the safe storage of OTC and prescribed medicines must be provided by prescribers, community pharmacies and other outlets of such medication.

Assessing the value of integrating national longitudinal shopping data into respiratory disease forecasting models

The COVID-19 pandemic led to unparalleled pressure on healthcare services. Improved healthcare planning in relation to diseases affecting the respiratory system has consequently become a key concern. We investigated the value of integrating sales of non-prescription medications commonly bought for managing respiratory symptoms, to improve forecasting of weekly registered deaths from respiratory disease at local levels across England, by using over 2 billion transactions logged by a UK high street retailer from March 2016 to March 2020. We report the results from the novel AI (Artificial Intelligence) explainability variable importance tool Model Class Reliance implemented on the PADRUS model (Prediction of Amount of Deaths by Respiratory disease Using Sales). PADRUS is a machine learning model optimised to predict registered deaths from respiratory disease in 314 local authority areas across England through the integration of shopping sales data and focused on purchases of non-prescription medications. We found strong evidence that models incorporating sales data significantly out-perform other models that solely use variables traditionally associated with respiratory disease (e.g. sociodemographics and weather data). Accuracy gains are highest (increases in R2 (coefficient of determination) between 0.09 to 0.11) in periods of maximum risk to the general public. Results demonstrate the potential to utilise sales data to monitor population health with information at a high level of geographic granularity.

Risk of Venous Thromboembolism With Gout Flares

OBJECTIVE

Previous studies demonstrated that the risk of venous thromboembolism (VTE) is increased in patients with gout, but not whether there was a temporal association between gout flare and VTE. This study was undertaken to evaluate potential temporal associations between gout flare and VTE.

METHODS

Data were obtained from electronic primary-care records from the UK's Clinical Practice Research Datalink, which links data from hospitalization and mortality registers. Using self-controlled case series analysis adjusted for season and age, we evaluated the temporal association between gout flare and VTE. The 90 days after primary-care consultation or hospitalization for gout flare was designated the exposed period. This was divided into three 30-day intervals. The baseline period was up to 2 years before the start of and up to 2 years after the end of the exposed period. The association between gout flare and VTE was measured using adjusted incidence rate ratios (IRRs) with 95% confidence intervals (95% CIs).

RESULTS

In total, 314 patients met the inclusion criteria (age ≥18 years, incident gout, no presence of VTE or use of a primary-care anticoagulant prescription before the start of the pre-exposure period). Among the 314 patients, VTE incidence was significantly higher in the exposed period than in the baseline period (adjusted IRR 1.83, 95% CI 1.30-2.59). The adjusted IRR of VTE during the first 30 days after gout flare was 2.31 (95% CI 1.39-3.82) relative to the baseline period. No increase in the adjusted IRRs was observed in days 31-60 (adjusted IRR 1.49, 95% CI 0.79-2.81) and days 61-90 (adjusted IRR 1.67, 95% CI 0.91-3.06) relative to baseline. Results were consistent across sensitivity analyses.

CONCLUSION

Among patients with gout, there was a transient increase in the rate of VTE within 30 days after primary-care consultation or hospitalization for gout flare.

Are people with mesothelioma who respond to the English Cancer Patient Experience Survey representative of the national mesothelioma population? A data comparison with cancer registry patients from the National Lung Cancer Audit

Introduction

More than 2,700 people are diagnosed with mesothelioma each year in the UK. Survival from mesothelioma is poor and to ensure service users' perspectives are incorporated in designing the most effective strategies to improve healthcare in England, the English Cancer Patient Experience Survey (CPES) is carried out annually in people diagnosed with cancer. We assessed whether the mesothelioma population responding to the English CPES is representative of the national mesothelioma population.

Method

Data from all people in England from the National Cancer Registry defined as having an incident mesothelioma diagnosis (2009-2015) were included. Data were linked across multiple sources including CPES. Using multivariable logistic regression, clinical and sociodemographic characteristics were compared between CPES respondents with mesothelioma and all people with mesothelioma in England.

Results

We identified 15,587 people diagnosed with mesothelioma, of which 1,597 (10.3%) were included in CPES. Adjusted odds ratios showed that representation in CPES decreased with older age, later stage, worse performance status, multiple comorbidities or emergency presentation at diagnosis. Gender was reasonably represented, although people with non-white ethnicity and from more deprived socioeconomic groups were underrepresented. lung cancer nurse specialist assessment was not associated with inclusion in CPES, however, having a lung cancer nurse specialist present at diagnosis was. Representation in CPES was highest for people who had chemo-radiotherapy (fully adjusted odds ratio 6.52 (95% confidence interval 5.34-7.96).

Conclusion

Responses were included across all sociodemographic and clinical groups, but there was unbalanced representation when compared with the national mesothelioma population. Patients who do not receive anti-cancer treatment are particularly under-represented. It is important to consider the impact of person characteristics on CPES representation when using cancer experiences to plan service management.

Association Between Gout Flare and Subsequent Cardiovascular Events Among Patients With Gout

IMPORTANCE

Gout is associated with cardiovascular diseases. The temporal association between gout flares and cardiovascular events has not been investigated.

OBJECTIVE

To investigate whether there is a transient increase in risk of cardiovascular events after a recent gout flare.

DESIGN, SETTING, AND PARTICIPANTS

A retrospective observational study was conducted using electronic health records from the Clinical Practice Research Datalink in England between January 1, 1997, and December 31, 2020. A multivariable nested case-control study was performed among 62 574 patients with gout, and a self-controlled case series, adjusted for season and age, was performed among 1421 patients with gout flare and cardiovascular event.

EXPOSURES

Gout flares were ascertained using hospitalization, primary care outpatient, and prescription records.

MAIN OUTCOMES AND MEASURES

The primary outcome was a cardiovascular event, defined as an acute myocardial infarction or stroke. Association with recent prior gout flares was measured using adjusted odds ratios (ORs) with 95% CIs in a nested case-control study and adjusted incidence rate ratios (IRRs) with 95% CIs in a self-controlled case series.

RESULTS

Among patients with a new diagnosis of gout (mean age, 76.5 years; 69.3% men, 30.7% women), 10 475 patients with subsequent cardiovascular events were matched with 52 099 patients without cardiovascular events. Patients with cardiovascular events, compared with those who did not have cardiovascular events, had significantly higher odds of gout flare within the prior 0 to 60 days (204/10 475 [2.0%] vs 743/52 099 [1.4%]; adjusted OR, 1.93 [95% CI, 1.57-2.38]) and within the prior 61 to 120 days (170/10 475 [1.6%] vs 628/52 099 [1.2%]; adjusted OR, 1.57 [95% CI, 1.26-1.96]). There was no significant difference in the odds of gout flare within the prior 121 to 180 days (148/10 475 [1.4%] vs 662/52 099 [1.3%]; adjusted OR, 1.06 [95% CI, 0.84-1.34]). In the self-controlled case series (N = 1421), cardiovascular event rates per 1000 person-days were 2.49 (95% CI, 2.16-2.82) within days 0 to 60; 2.16 (95% CI, 1.85-2.47) within days 61 to 120; and 1.70 (95% CI, 1.42-1.98) within days 121 to 180 after a gout flare, compared with cardiovascular event rates of 1.32 (95% CI, 1.23-1.41) per 1000 person-days within the 150 days before or the 181 to 540 days after the gout flare. Compared with 150 days before or the 181 to 540 days after a gout flare, incidence rate differences for cardiovascular events were 1.17 (95% CI, 0.83-1.52) per 1000 person-days, and adjusted IRRs were 1.89 (95% CI, 1.54-2.30) within days 0 to 60; 0.84 (95% CI, 0.52-1.17) per 1000 person-days and 1.64 (95% CI, 1.45-1.86) within days 61 to 120; and 0.38 (95% CI, 0.09-0.67) per 1000 person-days and 1.29 (95% CI, 1.02-1.64) within days 121 to 180 after a gout flare.

CONCLUSIONS AND RELEVANCE

Among individuals with gout, those who experienced a cardiovascular event, compared with those who did not experience such an event, had significantly higher odds of a recent gout flare in the preceding days. These findings suggest gout flares are associated with a transient increase in cardiovascular events following the flare.

Using Shopping Data to Improve the Diagnosis of Ovarian Cancer: Survey Study (Preprint)

BACKGROUND

Shopping data can be analyzed using machine learning techniques to study population health. It is unknown if the use of such methods can successfully investigate prediagnosis purchases linked to self-medication of symptoms of ovarian cancer.

OBJECTIVE

The aims of this study were to gain new domain knowledge from women's experiences, understand how women's shopping behavior relates to their pathway to the diagnosis of ovarian cancer, and inform research on computational analysis of shopping data for population health.

METHODS

A web-based survey on individuals' shopping patterns prior to an ovarian cancer diagnosis was analyzed to identify key knowledge about health care purchases. Logistic regression and random forest models were employed to statistically examine how products linked to potential symptoms related to presentation to health care and timing of diagnosis.

RESULTS

Of the 101 women surveyed with ovarian cancer, 58.4% (59/101) bought nonprescription health care products for up to more than a year prior to diagnosis, including pain relief and abdominal products. General practitioner advice was the primary reason for the purchases (23/59, 39%), with 51% (30/59) occurring due to a participant's doctor believing their health problems were due to a condition other than ovarian cancer. Associations were shown between purchases made because a participant's doctor believing their health problems were due to a condition other than ovarian cancer and the following variables: health problems for longer than a year prior to diagnosis (odds ratio [OR] 7.33, 95% CI 1.58-33.97), buying health care products for more than 6 months to a year (OR 3.82, 95% CI 1.04-13.98) or for more than a year (OR 7.64, 95% CI 1.38-42.33), and the number of health care product types purchased (OR 1.54, 95% CI 1.13-2.11). Purchasing patterns are shown to be potentially predictive of a participant's doctor thinking their health problems were due to some condition other than ovarian cancer, with nested cross-validation of random forest classification models achieving an overall in-sample accuracy score of 89.1% and an out-of-sample score of 70.1%.

CONCLUSIONS

Women in the survey were 7 times more likely to have had a duration of more than a year of health problems prior to a diagnosis of ovarian cancer if they were self-medicating based on advice from a doctor rather than having made the decision to self-medicate independently. Predictive modelling indicates that women in such situations, who are self-medicating because their doctor believes their health problems may be due to a condition other than ovarian cancer, exhibit distinct shopping behaviors that may be identifiable within purchasing data. Through exploratory research combining women sharing their behaviors prior to diagnosis and computational analysis of these data, this study demonstrates that women's shopping data could potentially be useful for early ovarian cancer detection.

Using patient experiences to evaluate care and expectations in lung cancer: analysis of the English Cancer Patient Experience Survey linked with the national cancer registry

Purpose

Identification of unmet needs in person centred and supportive care could be limited by differences in experience across specific cancer populations. Using the experiences of people with lung cancer, we assess distinctions according to demographic and clinical characteristics.

Methods

The English Cancer Patient Experience Survey was linked to the national cancer registry. The primary outcome was experience of the lung cancer pathway when assessed in multi-question models developed with item response theory. Secondary outcomes were experience by treatment received and in separate dimensions of the care pathway: up to diagnosis, treatment information, and staff support.

Results

Responses from 15,967 adults with a lung cancer diagnosis between 2009 and 2015 were included. Positive experiences were more likely to be reported by people aged between 65 and 80 (adjusted coefficient 0.08, 95%CI 0.05;0.11), those living in the most deprived areas (adjusted coefficient 0.10, 95%CI 0.05;0.14), diagnosed at lung cancer stage IIA–B (adjusted coefficient 0.09, 95%CI 0.04;0.14), and those diagnosed through inpatient elective admissions (adjusted coefficient 0.17, 95%CI 0.07;0.28). Specific experiences differed across dimensions of care and within lung cancer treatment groups.

Conclusions

Experiences differed according to gender and ethnicity, supporting previous observations in cancer. In contrast to previous studies, people with lung cancer were more likely to report positive pathway experiences at older ages, living in more deprived areas, or diagnosed after stage I, all frequently associated with worse clinical outcomes. The distinct observations in lung cancer specific analyses suggest potential unmet needs, such as in early stage disease and younger age groups.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00520-022-06863-4.

Access to a piped water supply is positively associated with haemoglobin levels in females living in rural Maharashtra State, India

OBJECTIVES

There is a high prevalence of anaemia in individuals living in rural India, which may be modified by a number of environmental factors. The association between access to water, toileting facilities and healthcare services with the prevalence of anaemia was explored to determine potentially modifiable community-level risk factors.

STUDY DESIGN

This was a cross-sectional survey.

METHODS

Data were collected from adolescent females (aged 13-17 years) living in 34 villages in rural areas of the Maharashtra state of India on measures of sanitation facilities and access to health care along with haemoglobin measurements. Linear and logistic regression analyses were conducted to investigate associations between environmental (community) factors and adolescent haemoglobin levels and anaemia, respectively.

RESULTS

Data were available from 1010 individuals, which represented a response rate of over 97% of those who were approached for the study. The prevalence of anaemia was very high (87%) when measured using haemoglobin levels. Access to a piped water supply was associated with 0.59 g/dL of increase in haemoglobin levels (95% confidence interval: 0.10-1.09). Associations between access to communal toilets, travel time to the hospital, health centres or nurses and haemoglobin levels or anaemia were not statistically significant.

CONCLUSIONS

Anaemia prevalence was very high in our study population. Simple improvements such as provision of regular piped water is associated with an increase in haemoglobin levels in rural Indian females. These are consistent with the hypothesis that chronic exposure to higher levels of microbes in the living environment contributes to the risk of anaemia.

External validation of a model to predict women most at risk of postpartum venous thromboembolism: Maternity clot risk

INTRODUCTION

Venous thromboembolism (VTE) is the leading cause of direct maternal mortality in high-income countries. We previously developed a risk prediction score for postpartum venous thromboembolism (VTE) in women without a previous VTE. In this paper, we provide further external validation and assess its performance across various groups of postpartum women from England.

MATERIALS AND METHODS

Cohort study using primary and secondary care data covering England. We used data from QResearch comprising women with pregnancies ending in live birth or stillbirth recoded in Hospital Episodes Statistics between 2004 and 2015. Outcome was VTE in the 6 weeks postpartum. Our predictor variables included sociodemographic and lifestyle characteristics, pre-existing comorbidities, and pregnancy and delivery characteristics.

RESULTS

Among 535,583 women with 700,185 deliveries, 549 VTE events were recorded (absolute risk of 7.8 VTE events per 10,000 deliveries). When we compared predicted probabilities of VTE for each woman from the original model with actual VTE events, we obtained a C-statistic of 0.67 (95% CI 0.65 to 0.70). However, our model slightly over-predicted VTE risk for the higher risk women (calibration slope = 0.84; 95% CI 0.74 to 0.94). Performance was similar across groups defined by calendar time, socioeconomic status, age group and geographical area. The score performed comparably with the existing algorithm used by the UK Royal College of Obstetrician and Gynaecologists.

CONCLUSIONS

Our model enables flexibility in setting new treatment thresholds. Adopting it in clinical practice may help optimise use of low-molecular-weight heparin postpartum to maximise health gain by better targeting of high-risk groups.

Two measures of systemic inflammation are positively associated with haemoglobin levels in adolescent girls living in rural India: a cross-sectional study

OBJECTIVE

This study tested the hypothesis that systemic inflammation is inversely associated with haemoglobin levels in adolescent girls in India.

METHODS

The study population consisted of adolescent girls aged between 10 and 19 years living in a remote rural region in Maharashtra State, India. Data were collected on anthropometric measures, and a venous blood sample was taken and tested for complete blood count and C-reactive protein (CRP).

RESULTS

Of 679 individuals who were invited to the research site to participate, data were available from 401 participants giving a response rate of 59%. Median blood CRP was 1.26 mg/l (Range 0.00 to 26.33), and 167 (41.6%) participants had CRP level < 1.0 mg/l. The mean haemoglobin was 12.24 g/dl (standard deviation [SD] 1.51), and the mean total white blood cells (WBC) count was 9.02 × 10³ /μl (SD 2.00). With each g/dl increase in blood haemoglobin, the risk of having an elevated CRP of ≥ 1 mg/l increased with an odds ratio of 1.16 (95% CI 1.01 to 1.33, P = 0.03). Total WBC count was also positively associated with blood haemoglobin, increasing by 0.24 × 10³ /μl (95% CI 0.11 to 0.37, P < 0.001) per g/dl increase in haemoglobin. Both analyses were adjusted for age.

CONCLUSIONS

In this population, blood haemoglobin levels were positively associated with two measures of systemic inflammation, contrary to the primary hypothesis being tested. Other unmeasured environmental exposures may modify haemoglobin levels in this population. Understanding this observation may help design better public health interventions to improve the well-being of adolescent girls in India.

Do working practices of cancer nurse specialists improve clinical outcomes? Retrospective cohort analysis from the English National Lung Cancer Audit

BACKGROUND

Cancer nurse specialists are advanced practitioners who offer continuity of care and expert support for people diagnosed with specific cancers. Health Education England's Cancer Workforce Plan prioritises expansion of cancer nurse specialist numbers by 2021 as part of the Cancer Taskforce Strategy for England.

OBJECTIVE

To assess whether working practices of advanced practice specialist nurses are associated with clinical outcomes for people with lung cancer.

METHODS

Adults with non-small cell lung cancer followed from 30 days post-diagnosis in English secondary care were obtained from the English National Lung Cancer Audit, 2007 to 2011. A national survey of lung cancer nurse specialists provided information on self-reported working practices. Mortality and unplanned admissions from 30 days to 12 months post diagnosis were respectively analysed using Cox and Poisson regression. Outcomes were assessed according to patients' receipt of initial assessments by a lung cancer nurse specialist and according to trust-level reported working practices. Regression models were adjusted for individual sociodemographic and clinical characteristics, error adjusted for intracorrelations within regional cancer networks, and presented separately according to patients' treatment pathways (surgery, chemotherapy, radiotherapy, or no anti-cancer therapy).

RESULTS

Data for 108,115 people with lung cancer were analysed and associations with mortality and unplanned admissions were infrequent. Among people receiving only radiotherapy, however, the hazard for death was 17% lower among those who received an assessment by a lung cancer nurse specialist, compared with no assessment (hazard ratio = 0.83, 95% confidence interval 0.73-0.94; p = 0.003). The hazard was also lower among those receiving surgery (hazard ratio = 0.91, 0.84-0.99; p = 0.028). Among those receiving radiotherapy, nurse specialists' reported confidence within multidisciplinary team settings was associated with a lower risk of death (hazard ratio = 0.88, 0.78-1.00; p = 0.049) and a lower rate of unplanned cancer-related admissions (incidence rate ratio = 0.83, 0.73-0.95; p = 0.007). Lung cancer nurse specialist assessments before/at diagnosis, were associated with a 5% lower rate of unplanned admissions, compared to when assessments occurred after diagnosis.

CONCLUSION

The contribution of nurse specialist working practices was occasionally associated with better outcomes for people with lung cancer. These were not limited to a single treatment pathway, but do indicate discrete relationships within pathways. Our study provides initial measures of overall lung cancer nurse specialist working practices at trusts, however, more detailed studies with longitudinal measurement of lung cancer nurse specialist-patient interaction are needed to better ascertain impacts on long-term patient outcomes. The findings highlight opportunities for potential improvement in effectiveness of service and care management.

Stillbirth Among Women Prescribed Nicotine Replacement Therapy in Pregnancy: Analysis of a Large UK Pregnancy Cohort

INTRODUCTION

We aimed to compare risk of stillbirth between maternal smokers and those prescribed nicotine replacement therapy (NRT) during pregnancy.

AIMS AND METHODS

We conducted a cross-sectional analysis on a pregnancy cohort of 220,630 singleton pregnancies ending in live or stillbirth between 2001 and 2012 from The Health Improvement Network UK general practice database. Women were categorized into three groups: NRT (prescribed during pregnancy or 1 month before conception); smokers; and controls (nonsmokers without a pregnancy NRT prescription). We calculated Odds ratios (OR) and corresponding 95% confidence intervals (CI) for stillbirth in the NRT group and smokers compared to controls.

RESULTS

A total of 805 pregnancies ended in stillbirth (3.6/1000 births). Absolute risks of stillbirth in NRT and smoker groups were both 5/1000 births compared with 3.5/1000 births in the control group. Compared with the control group, the adjusted odds of stillbirth in the NRT group was not statistically significant (OR = 1.35, 95% CI 0.91 to 2.00), although it was similar in magnitude to that in the smokers group (OR = 1.41, 95% CI 1.13 to 1.77).

CONCLUSIONS

We found no evidence of a statistically significant association between being prescribed NRT during pregnancy and odds of stillbirth compared with nonsmoking women. Although our study had much larger numbers than any previously, an even larger study with biochemically validated smoking outcome data and close monitoring of NRT use throughout pregnancy is required to exclude effects on findings of potential exposure misclassification.

Weight, height, and midupper arm circumference are associated with haemoglobin levels in adolescent girls living in rural India: A cross-sectional study

We aimed to explore the association of physical parameters with haemoglobin (Hb) levels to test the hypothesis that impaired physical development is associated with anaemia. A cross‐sectional survey study recruited adolescent girls (13 to 17 years) living in rural areas of Maharashtra state of India. Data were collected on physical parameters include height, weight, and midupper arm circumference (MUAC). Hb levels were measured using Sahli's haemometer. Linear regression was conducted to test the hypothesis. Data were collected from 1,010 girls on physical parameter and Hb levels. The majority of the adolescent girls were diagnosed with anaemia (87%). The regression analysis adjusted for age gave a significant association of Hb levels with all three variables (MUAC, weight, and height). Hb increased by 0.11 g/dl with an each centimetre of increase in MUAC (95% confidence interval, CI, [0.08, 0.15], P < .001). Each kilogram of increase in the body weight showed an increase in Hb levels (0.02 g dl, 95% CI [0.01, 0.03], P = .001). With an each centimetre of increase in height, Hb increased by 0.01 g dl (95% CI [0.00, 0.02], P = .022). There was a consistent association between three measures of somatic growth and anaemia in the study population. It is likely that life‐course exposures from conception onwards contribute to this, and the public health implications are that preventing anaemia is a challenge that requires a multifaceted interventional approach. Understanding the importance of the timing of these life exposures will help design interventions that can achieve optimal results.

Clinical management of nausea and vomiting in pregnancy and hyperemesis gravidarum across primary and secondary care: a population-based study

OBJECTIVES

To assess how nausea and vomiting in pregnancy (NVP) and hyperemesis gravidarum (HG) are managed and treated across primary and secondary care.

DESIGN

Population-based pregnancy cohort.

SETTING

Medical records (CPRD-GOLD) from England.

POPULATION

417 028 pregnancies during 1998-2014.

METHODS

Proportions of pregnancies with recorded NVP/HG diagnoses, primary care treatment, and hospital admissions were calculated. Multinomial logistic regression was employed to estimate adjusted relative risk ratios (aRRRs) with 99% confidence intervals (CIs) for the association between NVP/HG management paths and maternal characteristics.

MAIN OUTCOME MEASURES

NVP/HG diagnoses, treatments, and hospital admissions.

RESULTS

Overall prevalence of clinically recorded NVP/HG was 9.1%: 2.1% had hospital admissions, 3.4% were treated with antiemetics in primary care only, and 3.6% had only recorded diagnoses. Hospital admissions and antiemetic prescribing increased continuously during 1998-2013 (trend P < 0.001). Younger age, deprivation, Black/Asian/mixed ethnicity, and multiple pregnancy were associated with NVP/HG generally across all levels, but associations were strongest for hospital admissions. Most comorbidities had patterns of association with NVP/HG levels. Among women with NVP/HG who had no hospital admissions, 49% were prescribed antiemetics, mainly from first-line treatment (21% prochlorperazine, 15% promethazine, 13% cyclizine) and metoclopramide (10%). Of those admitted, 38% had prior antiemetic prescriptions (34% first-line, 9% second-line, 1% third-line treatment).

CONCLUSION

Previous focus on hospital admissions has greatly underestimated the NVP/HG burden. Although primary care prescribing has increased, most women admitted to hospital have no antiemetics prescribed before this. An urgent call is made to assess whether admissions could be prevented with better primary care recognition and timely treatment.

TWEETABLE ABSTRACT

The NVP/HG burden is increasing over time and management optimisation should be high priority to help reduce hospital admissions.

Assessment of discharge treatment prescribed to women admitted to hospital for hyperemesis gravidarum

AIMS

Prescribing drug treatment for the management of hyperemesis gravidarum (HG), the most severe form of nausea and vomiting in pregnancy, remains controversial. Since most manufacturers do not recommend prescribing antiemetics during pregnancy, little is known regarding which treatments are most prevalent among pregnant patients. Here, we report for the first time, evidence of actual treatments prescribed in English hospitals.

METHODS

A retrospective pregnancy cohort was constructed using anonymised electronic records in the Nottingham University Hospitals Trust system for all women who delivered between January 2010 and February 2015. For women admitted to hospital for HG, medications prescribed on discharge were described and variation by maternal characteristics was assessed. Compliance with local and national HG treatment guidelines was evaluated.

RESULTS

Of 33 567 pregnancies (among 30 439 women), the prevalence of HG was 1.7%. Among 530 HG admissions with records of discharge drugs, cyclizine was the most frequently prescribed (almost 73% of admissions). Prochlorperazine and metoclopramide were prescribed mainly in combination with other drugs; however, ondansetron was more common than metoclopramide at discharge from first and subsequent admissions. Steroids were only prescribed following readmissions. Thiamine was most frequently prescribed following readmission while high dose of folic acid was prescribed equally after first or subsequent admissions. Prescribing showed little variation by maternal age, ethnicity, weight, socioeconomic deprivation, or comorbidities.

CONCLUSION

Evidence that management of HG in terms of discharge medications mainly followed local and national recommendations provides reassurance within the health professional community. Wider documentation of drugs prescribed to women with HG is required to enable full assessment of whether optimal drug management is being achieved.

Estimating the health-care costs of children born to pregnant smokers in England: cohort study using primary and secondary health-care data

BACKGROUND AND AIMS

Little is known about the long-term economic consequences of smoking during pregnancy. We estimated the association between smoking in pregnancy and the costs of delivering health-care to infants and children in England, and investigated which aspects of care are the key drivers of these costs.

METHODS

We used Hospital Episode Statistics (HES) linked with Clinical Practice Research Datalink (CPRD) data in England from January 2003 to January 2015 in children with longitudinal data for at least 1, 5 and 10 years after birth. Poisson regression provided rate ratios (RR) and 95% confidence intervals (CIs) comparing health-care episode rates between those exposed and not exposed to smoking during pregnancy. Linear regression was used to compare estimated costs between groups (£ sterling, 2015 prices) and generalized linear multivariable (GLM) models adjusted for potentially moderating factors.

RESULTS

A total of 93 152 singleton pregnancies with the required data were identified. Maternal smoking in pregnancy was associated with higher primary care, prescription and hospital in-patient episode rates, but lower out-patient visit and diagnostic test rates. Adjusting for year of birth, socio-economic deprivation, parity, sex of child and delivery method showed that maternal smoking in pregnancy was associated with increased child health-care costs at 1 year [average cost difference for children of smokers, β = £91.18, 95% confidence interval (CI) = £47.52-134.83 and 5 years of age (β = £221.80, 95% CI = £17.78-425.83], but not at 10 years of age (β = £365.94, 95% CI = -£192.72 to £924.60).

CONCLUSION

In England, maternal smoking in pregnancy is associated with increased child health-care costs over the first 5 years of life; these costs are driven primarily by greater hospital in-patient care.

Adverse Maternal and Birth Outcomes in Women Admitted to Hospital for Hyperemesis Gravidarum: a Population-Based Cohort Study

BACKGROUND

Evidence for risks of adverse maternal and birth outcomes in women with hyperemesis gravidarum (HG) is predominantly from small studies, unknown, or conflicting.

METHODS

A population-based cohort study using secondary health care records (Hospital Episode Statistics covering all of England from 1997 to 2012) was used to calculate odds ratios (OR) with 99% confidence intervals (CI) for the association between HG hospital admission and adverse outcomes, adjusting for maternal and pregnancy confounders.

RESULTS

Within 8 211 850 pregnancies ending in live births or stillbirths, women with HG had increased odds of anaemia (OR 1.28, 99% CI 1.23, 1.33), preeclampsia (OR 1.16, 99% CI 1.09, 1.22), eclampsia (OR 1.84, 99% CI 1.07, 3.18), venous thromboembolism antenatally (OR 1.94, 99% CI 1.57, 2.39 for deep vein thrombosis, and OR 2.54, 99% CI 1.89, 3.40 for pulmonary embolism) and post-partum. Odds of stillbirth (OR 0.77, 99% CI 0.66, 0.89) and post-term (OR 0.86, 99% CI 0.81, 0.92) delivery were decreased. Women were more likely to be induced (OR 1.20, 99% CI 1.16, 1.23), to deliver preterm (OR 1.11, 99% CI 1.05, 1.17), very preterm (OR 1.18, 99% CI 1.05, 1.32), or by caesarean section (OR 1.12, 99% CI 1.08, 1.16), to have low birthweight (OR 1.12, 99% CI 1.08, 1.17) or small for gestational age (OR 1.06, 99% CI 1.01, 1.11) babies and although absolute risks were small, their offspring were more likely to undergo resuscitation or neonatal intensive care.

CONCLUSION

HG may have important antenatal and postnatal consequences that should be considered in communications between health care professionals and women to best manage HG and prevent progression during pregnancy.

Direct to Public Peer Support and e-Therapy Program Versus Information to Aid Self-Management of Depression and Anxiety: Protocol for a Randomized Controlled Trial

Background

Regardless of geography or income, effective help for depression and anxiety only reaches a small proportion of those who might benefit from it. The scale of the problem suggests a role for effective, safe, anonymized public health–driven Web-based services such as Big White Wall (BWW), which offer immediate peer support at low cost.

Objective

Using Reach, Effectiveness, Adoption, Implementation and Maintenance (RE-AIM) methodology, the aim of this study was to determine the population reach, effectiveness, cost-effectiveness, and barriers and drivers to implementation of BWW compared with Web-based information compiled by UK’s National Health Service (NHS, NHS Choices Moodzone) in people with probable mild to moderate depression and anxiety disorder.

Methods

A pragmatic, parallel-group, single-blind randomized controlled trial (RCT) is being conducted using a fully automated trial website in which eligible participants are randomized to receive either 6 months access to BWW or signposted to the NHS Moodzone site. The recruitment of 2200 people to the study will be facilitated by a public health engagement campaign involving general marketing and social media, primary care clinical champions, health care staff, large employers, and third sector groups. People will refer themselves to the study and will be eligible if they are older than 16 years, have probable mild to moderate depression or anxiety disorders, and have access to the Internet.

Results

The primary outcome will be the Warwick-Edinburgh Mental Well-Being Scale at 6 weeks. We will also explore the reach, maintenance, cost-effectiveness, and barriers and drivers to implementation and possible mechanisms of actions using a range of qualitative and quantitative methods.

Conclusions

This will be the first fully digital trial of a direct to public online peer support program for common mental disorders. The potential advantages of adding this to current NHS mental health services and the challenges of designing a public health campaign and RCT of two digital interventions using a fully automated digital enrollment and data collection process are considered for people with depression and anxiety.

Trial Registration

International Standard Randomized Controlled Trial Number (ISRCTN): 12673428; http://www.controlled-trials.com/ISRCTN12673428/12673428 (Archived by WebCite at http://www.webcitation.org/6uw6ZJk5a)

Peripartum hysterectomy: an economic analysis of direct healthcare costs using routinely collected data

OBJECTIVE

To estimate resource use and costs associated with peripartum hysterectomy for the English National Health Service.

DESIGN/SETTING

Analysis of linked Clinical Practice Research Datalink and Hospital Episodes Statistics (CPRD-HES) data.

POPULATION

Women undergoing peripartum hysterectomy between 1997 and 2013 and matched controls.

METHODS

Inverse probability weighted generalised estimating equations were used to model the non-linear trend in healthcare service use and costs over time, accounting for missing data, adjusting for maternal age, body mass index, delivery year, smoking and socio-economic indicators.

MAIN OUTCOME MEASURES

Primary care, hospital outpatient and inpatient attendances and costs (UK 2015 prices).

RESULTS

The study sample included 1362 women (192 cases and 1170 controls) who gave birth between 1997 and 2013; 1088 (153 cases and 935 controls) of these were deliveries between 2003 and 2013 when all categories of hospital resource use were available. Based on the 2003-2013 delivery cohort, peripartum hysterectomy was associated with a mean adjusted additional total cost of £5380 (95% CI £4436-6687) and a cost ratio of 1.76 (95% CI 1.61-1.98) over 5 years of follow up compared with controls. Inpatient costs, mostly incurred during the first year following surgery, accounted for 78% excluding or 92% including delivery-related costs.

CONCLUSION

Peripartum hysterectomy is associated with increased healthcare costs driven largely by increased post-surgery hospitalisation rates. To reduce healthcare costs and improve outcomes for women who undergo hysterectomy, interventions that reduce avoidable repeat hospitalisations following surgery such as providing active follow up, treatment and support in the community should be considered.

TWEETABLE ABSTRACT

A large amount of NHS data on peripartum hysterectomy suggests active community follow up could reduce costs, #HealthEconomics.

Risk factors for maternal anaemia and low birth weight in pregnant women living in rural India: a prospective cohort study

OBJECTIVE

The aim of this prospective study was to estimate the prevalence and risk factors for maternal anaemia and low birth weight (LBW) in pregnant women living in Maharashtra state, India.

STUDY DESIGN

This is a prospective study.

METHODS

Women between 3 and 5 months of pregnancy were recruited from 34 villages based in Maharashtra state. Baseline data collection, anthropometric measurements and blood investigations were performed. Participants were followed-up to record birth weight.

RESULTS

In total, 303 women were eligible, and 287 (95%) provided data. 77% were anaemic, defined as haemoglobin less than 11.0 g/dl at the time of recruitment, with a mean corpuscular volume of 80.5 fl/cell (standard deviation: 7.22, range: 53.4-93.8). The increased risk of anaemia was seen in women with consanguineous marriages (odds ratio [OR]: 2.41, 95% confidence interval [CI]: 1.16-5.01, P = 0.01) after adjustment for potential confounding factors. Postdelivery data from full-term singleton live births demonstrated a 7% prevalence of LBW. Consanguineous marriage was a major risk factor for LBW (OR: 4.10, 95% CI: 1.25-13.41, P = 0.02). The presence of maternal anaemia during 3-5 months of pregnancy was associated with lower risk of LBW (unadjusted OR: 0.34, 95% CI: 0.13-0.92, P = 0.03).

CONCLUSION

About 30% of our study participants were in a consanguineous marriage, which was identified as a potentially avoidable risk factor for both anaemia and LBW.

Incidence of First Stroke in Pregnant and Nonpregnant Women of Childbearing Age: A Population-Based Cohort Study From England

Background

Pregnant women may have an increased risk of stroke compared with nonpregnant women of similar age, but the magnitude and the timing of such risk are unclear. We examined the risk of a first stroke event in women of childbearing age and compared the risk during pregnancy and in the early postpartum period with the background risk outside these periods.

Methods and Results

We conducted an open cohort study of 2 046 048 women aged 15 to 49 years between April 1, 1997, and March 31, 2014, using linked primary (Clinical Practice Research Datalink) and secondary (Hospital Episode Statistics) care records in England. Risk of first stroke was assessed by calculating the incidence rate of stroke in antepartum, peripartum (2 days before until 1 day after delivery), and early (first 6 weeks) and late (second 6 weeks) postpartum periods compared with nonpregnant time using a Poisson regression model with adjustment for maternal age, socioeconomic group, and calendar time. A total of 2511 women had a first stroke. The incidence rate of stroke was 25.0 per 100 000 person‐years (95% CI 24.0–26.0) in nonpregnant time. The rate was lower antepartum (10.7 per 100 000 person‐years, 95% CI 7.6–15.1) but 9‐fold higher peripartum (161.1 per 100 000 person‐years, 95% CI 80.6–322.1) and 3‐fold higher early postpartum (47.1 per 100 000 person‐years, 95% CI 31.3–70.9). Rates of ischemic and hemorrhagic stroke both increased peripartum and early postpartum.

Conclusions

Although the absolute risk of first stroke is low in women of childbearing age, healthcare professionals should be aware of a considerable increase in relative risk during the peripartum and early postpartum periods.

The incidence of first stroke in and around pregnancy: A population-based cohort study from Sweden

Introduction

Research has suggested that delivery is associated with an increased risk of stroke in women; however, there is a lack of contemporary estimates on the incidence of stroke in and after pregnancy compared with the baseline risk in women of childbearing age in Sweden.

Patients and methods

All women aged 15-49 years with live births/stillbirths in 1992-2011 were identified from the Swedish Medical Birth Registry linked with the National Patient Registry. First stroke during the study period was identified. Incidence rates per 100,000 person-years and adjusted incidence rate ratios (IRRs) were calculated for antepartum, peripartum and early and late postpartum periods, compared with all other available follow-up time (time before pregnancy and after postpartum) using Poisson regression adjusted for maternal age, education attainment and calendar time.

Results

Of 1,124,541 women, 3094 had a first incident stroke (331 occurred during pregnancy or first 12 weeks postpartum), about half having ischaemic stroke. The incidence was 15.0 per 100,000 person-years (95% confidence interval 14.5-15.6) in non-pregnant time. The incidence was lower antepartum (7.3/100,000 person-years, 6.0-8.9; adjusted IRR = 0.7, 0.5-0.8) but higher peripartum (314.4/100,000 person-years, 247.5-399.5; adjusted IRR = 27.3, 21.4-34.9) and early postpartum (64.0/100,000 person-years, 54.1-75.7; adjusted IRR = 5.5, 4.6-6.6). The increased risk in peripartum was more evident for intracerebral haemorrhage than other types of stroke.

Conclusion

Overall risk of stroke was low in women of childbearing age, but stroke risk peaks in the peripartum and early postpartum periods. Future work should address factors that contribute to this increased risk in order to develop approaches to attenuate risk.

Development and validation of risk prediction model for venous thromboembolism in postpartum women: multinational cohort study

OBJECTIVE

 To develop and validate a risk prediction model for venous thromboembolism in the first six weeks after delivery (early postpartum).

DESIGN

 Cohort study.

SETTING

 Records from England based Clinical Practice Research Datalink (CPRD) linked to Hospital Episode Statistics (HES) and data from Sweden based registry.

PARTICIPANTS

 All pregnant women registered with CPRD-HES linked data between 1997 and 2014 and Swedish medical birth registry between 2005 and 2011 with postpartum follow-up.

MAIN OUTCOME MEASURE

 Multivariable logistic regression analysis was used to develop a risk prediction model for postpartum venous thromboembolism based on the English data, which was externally validated in the Swedish data.

RESULTS

 433 353 deliveries were identified in the English cohort and 662 387 in the Swedish cohort. The absolute rate of venous thromboembolism was 7.2 per 10 000 deliveries in the English cohort and 7.9 per 10 000 in the Swedish cohort. Emergency caesarean delivery, stillbirth, varicose veins, pre-eclampsia/eclampsia, postpartum infection, and comorbidities were the strongest predictors of venous thromboembolism in the final multivariable model. Discrimination of the model was similar in both cohorts, with a C statistic above 0.70, with excellent calibration of observed and predicted risks. The model identified more venous thromboembolism events than the existing national English (sensitivity 68% v 63%) and Swedish guidelines (30% v 21%) at similar thresholds.

CONCLUSION

 A new prediction model that quantifies absolute risk of postpartum venous thromboembolism has been developed and externally validated. It is based on clinical variables that are available in many developed countries at the point of delivery and could serve as the basis for real time decisions on obstetric thromboprophylaxis.

Differing patterns in thermal injury incidence and hospitalisations among 0-4 year old children from England

OBJECTIVE

To describe patterns in thermal injury incidence and hospitalisations by age, gender, calendar year and socioeconomic status among 0-4 year olds in England for the period 1998-2013.

PARTICIPANTS

708,050 children with linked primary care and hospitalisation data from the Clinical Practice Research Datalink (CPRD) and Hospital Episode Statistics (HES), respectively.

ANALYSIS

Incidence rates of all thermal injuries (identified in CPRD and/or HES), hospitalised thermal injuries, and serious thermal injuries (hospitalised for ≥72h). Adjusted incidence rate ratios (IRR) and 95% confidence intervals (95%CI), estimated using Poisson regression.

RESULTS

Incidence rates of all thermal injuries, hospitalised thermal injuries, and serious thermal injuries were 59.5 per 10,000 person-years (95%CI 58.4-60.6), 11.3 (10.8-11.8) and 2.15 (1.95-2.37), respectively. Socioeconomic gradients, between the most and least deprived quintiles, were steepest for serious thermal injuries (IRR 3.17, 95%CI 2.53-3.96). Incidence of all thermal injuries (IRR 0.64, 95%CI 0.58-0.70) and serious thermal injuries (IRR 0.44, 95%CI 0.33-0.59) reduced between 1998/9 and 2012/13. Incidence rates of hospitalised thermal injuries did not significantly change over time.

CONCLUSION

Incidence of all thermal injuries and those hospitalised for ≥72h reduced over time. Steep socioeconomic gradients support continued targeting of preventative interventions to those living in the most deprived areas.

Prevalence of iron-deficiency anaemia and risk factors in 1010 adolescent girls from rural Maharashtra, India: a cross-sectional survey

OBJECTIVE

Iron-deficiency anaemia (IDA) is the most common nutritional disorder observed in adolescent girls in India. Our aim was to investigate the prevalence and risk factors associated with IDA in rural Maharashtra, India, to address current evidence gaps.

STUDY DESIGN

Cross-sectional survey.

METHODS

The study recruited 13- to 17-year-old adolescent girls living in 34 villages of Osmanabad district. Data were collected on individual health, dietary, sociodemographic factors, and anthropometric measurements were taken. Haemoglobin (Hb) levels were measured using Sahli's haemometer. Logistic and linear regressions were used to identify risk factors associated with IDA and Hb levels, respectively.

RESULTS

Among 1010 adolescent girls (response rate 97.5%), the mean Hb was 10.1 g/dl (standard deviation = 1.3), and 87% had anaemia (Hb < 12 g/dl). The prevalence of mild (11.0-11.9 g/dl), moderate (8.0-10.9 g/dl) and severe (Hb ≤ 7.9 g/dl) anaemia was 17%, 65% and 5%, respectively. Anaemia likelihood increased significantly with age (odds ratio (OR): 1.41 per year, 95% confidence interval (CI): 1.17-1.70). Factors associated with decreased anaemia risk were mid-upper arm circumference (MUAC) ≥22 cm (OR: 0.51, 95% CI: 0.31-0.82), ≥3 days/week consumption of fruit (OR: 0.35, 95% CI: 0.23-0.54) or rice (OR: 0.39, 95% CI: 0.17-0.91), and incomplete schooling (OR: 0.47, 95% CI: 0.24-0.91). In the final model lower age, MUAC and fruit consumption were significantly associated with Hb level.

CONCLUSION

Anaemia prevalence was extremely high among adolescent girls in rural areas of Maharashtra. Whilst we identified risk factors that could be used for targeting interventions, there is urgent need of comprehensive preventative interventions for the whole adolescent girl population.

The risk of community-acquired pneumonia among 9803 patients with coeliac disease compared to the general population: a cohort study

BACKGROUND

Patients with coeliac disease are considered as individuals for whom pneumococcal vaccination is advocated.

AIM

To quantify the risk of community-acquired pneumonia among patients with coeliac disease, assessing whether vaccination against streptococcal pneumonia modified this risk.

METHODS

We identified all patients with coeliac disease within the Clinical Practice Research Datalink linked with English Hospital Episodes Statistics between April 1997 and March 2011 and up to 10 controls per patient with coeliac disease frequency matched in 10-year age bands. Absolute rates of community-acquired pneumonia were calculated for patients with coeliac disease compared to controls stratified by vaccination status and time of diagnosis using Cox regression in terms of adjusted hazard ratios (HR).

RESULTS

Among 9803 patients with coeliac disease and 101 755 controls, respectively, there were 179 and 1864 first community-acquired pneumonia events. Overall absolute rate of pneumonia was similar in patients with coeliac disease and controls: 3.42 and 3.12 per 1000 person-years respectively (HR 1.07, 95% CI 0.91-1.24). However, we found a 28% increased risk of pneumonia in coeliac disease unvaccinated subjects compared to unvaccinated controls (HR 1.28, 95% CI 1.02-1.60). This increased risk was limited to those younger than 65, was highest around the time of diagnosis and was maintained for more than 5 years after diagnosis. Only 26.6% underwent vaccination after their coeliac disease diagnosis.

CONCLUSIONS

Unvaccinated patients with coeliac disease under the age of 65 have an excess risk of community-acquired pneumonia that was not found in vaccinated patients with coeliac disease. As only a minority of patients with coeliac disease are being vaccinated there is a missed opportunity to intervene to protect these patients from pneumonia.

Epidemiology of poisonings, fractures and burns among 0-24 year olds in England using linked health and mortality data

BACKGROUND

Understanding patterns of injury in England is challenging due to a lack of national injury surveillance data. Through recent linkage of a large primary care research database to hospitalization and mortality data, we describe the epidemiology of poisonings, fractures and burns over a 14-year period.

METHODS

We used linked English primary care, hospitalisation and mortality data from the Clinical Practice Research Datalink, Hospital Episode Statistics and Office for National Statistics between 1998 and 2011 to establish a cohort of 2,106,420 0-24 year olds. Incidence rates, per 10 000 person-years (PY) were estimated by age, sex, calendar year and socioeconomic status. Using Poisson regression we estimated incidence rate ratios, adjusting for age and sex.

RESULTS

Age patterns of injury incidence varied by injury type, with peaks at age 2 (74.3/10 000 PY) and 18 (74.7/10 000 PY) for poisonings, age 13 for fractures (305.1/10 000 PY) and age 1 for burns (116.8/10 000 PY). Over time, fracture incidence increased, whereas poisoning incidence increased only among 15-24 year olds and burns incidence reduced. Poisoning and burns incidence increased with deprivation, with the steepest socioeconomic gradient for poisonings among 20-24 year olds (IRR 2.63, 95% confidence interval 2.24-3.09).

CONCLUSION

Differing patterns according to age and injury type reflect differences in underlying injury mechanisms, highlighting the importance of developing tailored preventative interventions across the life course. Inequalities in injury occurrences support the targeting of preventative interventions to children and young people living in the most deprived areas.

Hospital admission for hyperemesis gravidarum: a nationwide study of occurrence, reoccurrence and risk factors among 8.2 million pregnancies

STUDY QUESTION

What are the maternal risk factors for hyperemesis gravidarum (HG) hospital admission, readmission and reoccurrence in a following pregnancy?

SUMMARY ANSWER

Young age, less socioeconomic deprivation, nulliparity, Asian or Black ethnicity, female fetus, multiple pregnancy, history of HG in a previous pregnancy, thyroid and parathyroid dysfunction, hypercholesterolemia and Type 1 diabetes are all risk factors for HG.

WHAT IS KNOWN ALREADY

Women with Black or Asian ethnicity, of young age, carrying multiple babies or singleton females, with Type 1 diabetes or with a history of HG were previously reported to be at higher risk of developing HG; however, most evidence is from small studies. Little is known about associations with other comorbidities and there is controversy over other risk factors such as parity. Estimates of HG prevalence vary and there is a little understanding of the risks of HG readmission in a current pregnancy and reoccurrence rates in subsequent pregnancies, all of which are needed for planning measures to reduce onset or worsening of the condition.

STUDY DESIGN, SIZE, DURATION

We performed a population-based cohort study of pregnancies ending in live births and stillbirths using prospectively recorded secondary care records (Hospital Episode Statistics) from England. We analysed those computerized and anonymized clinical records from over 5.3 million women who had one or more pregnancies between 1997 and 2012.

PARTICIPANTS/MATERIALS, SETTING, METHODS

We obtained 8 215 538 pregnancies from 5 329 101 women of reproductive age, with a total of 186 800 HG admissions occurring during 121 885 pregnancies. Multivariate logistic regression with generalized estimating equations was employed to estimate odds ratios (aOR) to assess sociodemographic, pregnancy and comorbidity risk factors for HG onset, HG readmission within a pregnancy and reoccurrence in a subsequent pregnancy.

MAIN RESULTS AND THE ROLE OF CHANCE

Being younger, from a less socioeconomically deprived status, of Asian or Black ethnicity, carrying a female fetus or having a multiple pregnancy all significantly increased HG and readmission risk but only ethnicity increased reoccurrence. Comorbidities most strongly associated with HG were parathyroid dysfunction (aOR = 3.83, 95% confidence interval 2.28-6.44), hypercholesterolemia (aOR = 2.54, 1.88-3.44), Type 1 diabetes (aOR = 1.95, 1.82-2.09), and thyroid dysfunction (aOR = 1.85, 1.74-1.96). History of HG was the strongest independent risk factor (aOR = 4.74, 4.46-5.05). Women with higher parity had a lower risk of HG compared with nulliparous women (aOR = 0.90, 0.89-0.91), which was not explained by women with HG curtailing further pregnancies.

LIMITATIONS, REASONS FOR CAUTION

Although this represents the largest population-based study worldwide on the topic, the results could have been biased by residual and unmeasured confounding considering that some potential important risk factors such as smoking, BMI or prenatal care could not be measured with these data. Underestimation of non-routinely screened comorbidities such as hypercholesterolemia or thyroid dysfunction could also be a cause of selection bias.

WIDER IMPLICATIONS OF THE FINDINGS

The estimated prevalence of 1.5% from our study was similar to the average prevalence reported in the literature and the representativeness of our data has been validated by comparison to national statistics. Also the prevalence of comorbidities was mostly similar to other studies estimating these in the UK and other developed countries. Women with Black or Asian ethnicity, of young age, carrying multiple babies or singleton females, with Type 1 diabetes or with history of HG were confirmed to be at higher risk of HG with an unprecedented higher statistical power. We showed for the first time that socioeconomic status interacts with maternal age, that hypercholesterolemia is a potential risk factor for HG and that carrying multiple females increases risk of hyperemesis compared with multiple males. We also provided robust evidence for the association of parity with HG. Earlier recognition and management of symptoms via gynaecology day-case units or general practitioner services can inform prevention and control of consequent hospital admissions.

STUDY FUNDING/COMPETING INTERESTS

The work was founded by The Rosetrees Trust and the Stoneygate Trust. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. C.N.-P. reports personal fees from Sanofi Aventis, Warner Chilcott, Leo Pharma, UCB and Falk, outside the submitted work and she is one of the co-developers of the RCOG Green Top Guideline on HG; all other authors did not report any potential conflicts of interest.

TRIAL REGISTRATION NUMBER

Not applicable.

Changes in poisonings among adolescents in the UK between 1992 and 2012: a population based cohort study

BACKGROUND

Poisonings are a common cause of morbidity and mortality among adolescents. Yet surveillance data indicating current incidence rates (IRs) and time trends are lacking, making policy development and service planning difficult. We utilised population based primary care data to estimate adolescent poisoning rates according to intent across the UK.

METHODS

A cohort study of 1 311 021 adolescents aged 10-17 years, between 1992 and 2012, was conducted using routine primary care data from The Health Improvement Network. IRs and adjusted IRRs with 95% CIs were calculated for all poisonings, intentional, unintentional, unknown intent and alcohol related poisonings, by age, sex, calendar time and socioeconomic deprivation.

RESULTS

Overall poisoning incidence increased by 27% from the period 1992-1996 to 2007-2012, with the largest increases in intentional poisonings among females aged 16-17 years (IR 391.4/100 000 person years (PY), CI 328.9 to 465.7 for age 17 years in 1992-1996; 767.0/100 000 PY, CI 719.5 to 817.7 in 2007-2012) and alcohol related poisonings in females aged 15-16 years (IR 65.7/100 000 PY, CI 43.3 to 99.8 rising to 130.0/100 000 PY, CI 110.0 to 150.0 for age 15 years). A strong socioeconomic gradient for all poisonings persisted over time, with higher rates among the more deprived (IRR 2.63, CI 2.41 to 2.88 for the most vs least deprived quintile in 2007-2012).

CONCLUSIONS

Adolescent poisonings, especially intentional poisonings, have increased substantially over time and remain associated with health inequalities. Social and psychological support for adolescents should be targeted at more deprived communities, and child and adolescent mental health and alcohol support service provision should be commissioned to reflect the changing need.

Assisted reproductive technology and perinatal outcomes: conventional versus discordant-sibling design

OBJECTIVE

To compare risks of adverse perinatal outcomes between assisted reproductive technology (ART) and naturally conceived singleton births using a dual design approach.

DESIGN

Discordant-sibling and conventional cross-sectional general population comparison.

SETTING

Not applicable.

PATIENT(S)

All singleton live births, conceived naturally or via ART.

INTERVENTION(S)

None.

MAIN OUTCOME MEASURE(S)

Birth weight, gestational age, low birth weight, preterm delivery, small for gestational age (SGA), low Apgar score.

RESULT(S)

A total of 32,762 (0.8%) of 3,896,242 singleton live births in the three states were conceived via ART. In 6,458 sibling pairs, ART-conceived singletons were 33 g lighter (adjusted β = -33.40, 95% confidence interval [CI], -48.60, -18.21) and born half a day sooner (β = -0.58, 95% CI, -1.02, -0.14) than singletons conceived naturally. The absolute risk of low birth weight and preterm birth was 6.8% and 9.7%, respectively, in the ART group and 4.9% and 7.9%, respectively, in the non-ART group. The odds of low birth weight were 33% higher (adjusted odds ratio [aOR] = 1.33; 95% CI, 1.13, 1.56) and 20% higher for preterm birth (aOR = 1.20; 95% CI, 1.07, 1.34). The odds of SGA and low Apgar score were not significantly different in both groups (aOR = 1.22; 95% CI, 0.88, 1.68; and aOR = 0.75; 95% CI, 0.54, 1.05, respectively). Results of conventional analyses were similar, although the magnitude of risk was higher for preterm birth (aOR, 1.51; 95% CI 1.46, 1.56).

CONCLUSION(S)

Despite some inflated risks in the general population comparison, ART remained associated with an increased likelihood of low birth weight and preterm birth when underlying maternal factors were kept constant using discordant-sibling comparison.

Abstract 140: Incidence of Stroke in and Around Pregnancy: A UK Population-based Cohort Study

Introduction: There are very few population-based studies on the incidence of stroke in women of childbearing age, stratifying by stroke types and pregnancy-related periods.

Methods: We used an open cohort study design including all women aged 15-49 years from UK linked primary (Clinical Practice Research Datalink) and secondary (Hospital Episode Statistics) care records in 1997-2014. The exposure of our study was pregnancy resulting in a live or a stillbirth and associated antenatal and postpartum periods. The outcome of the study was the first ever stroke diagnosis, defined using ICD-10 codes (I60-I64, O22.5 and O87.3) or relevant Read codes, and classified as having ischaemic stroke (IS), intracerebral haemorrhage (ICH), subarachnoid haemorrhage (SAH) or unspecified. We calculated the absolute rates of stroke per 100,000 person-years and 95% confidence intervals (95%CI) for different exposure periods. We stratified the analysis by maternal age and types of stroke.

Results: Of 2,047,858 women, we identified 336,957 women with 453,776 deliveries. There were totally 2,526 women with a first incidence of stroke: IS 1,140 (45.1%), SAH 684 (27.1%), ICH 368 (14.6%) and unspecified 334 (13.2%). The overall incidence rate of stroke was 24.9 (95%CI 23.6-26.2) per 100,000 person-years in the non-pregnant period (IS 11.2 [10.4-12.1], ICH 3.6 [3.2-4.2], SAH 6.8 [6.2-7.5] and unspecified 3.3 [2.9-3.8]). The incidence was however higher around delivery (281.9 [141.6-561.2]) and in the first six weeks postpartum (43.8 [25.3-75.9]) and the rate ratios compared to the non-pregnant period after adjusting for age were 19.2 (9.6-38.3) and 3.0 (1.7-5.2) respectively.

Conclusions: Although the incidence of stroke for young women was relatively low, the incidence around delivery or in the early postpartum was significantly higher compared to other periods, regardless of maternal age.

Defining venous thromboembolism and measuring its incidence using Swedish health registries: a nationwide pregnancy cohort study

OBJECTIVE

To accurately define venous thromboembolism (VTE) in the routinely collected Swedish health registers and quantify its incidence in and around pregnancy.

STUDY DESIGN

Cohort study using data from the Swedish Medical Birth Registry (MBR) linked to the National Patient Registry (NPR) and the Swedish Prescribed Drug Register (PDR).

SETTING

Secondary care centres, Sweden.

PARTICIPANT

509,198 women aged 15-44 years who had one or more pregnancies resulting in a live birth or stillbirth between 2005 and 2011.

MAIN OUTCOME MEASURE

To estimate the incidence rate (IR) of VTE in and around pregnancy using various VTE definitions allowing direct comparison with other countries.

RESULTS

The rate of VTE varied based on the VTE definition. We found that 43% of cases first recorded as outpatient were not accompanied by anticoagulant prescriptions, whereas this proportion was much lower than those cases first recorded in the inpatient register (9%). Using our most inclusive VTE definition, we observed higher rates of VTE compared with previously published data using similar methodology. These reduced by 31% (IR=142/100,000 person-years; 95% CI 132 to 153) and 22% (IR=331/100,000 person-years; 95% CI 304 to 361) during the antepartum and postpartum periods, respectively, using a restrictive VTE definition that required anticoagulant prescriptions associated with diagnosis, which were more in line with the existing literature.

CONCLUSIONS

We found that including VTE codes without treatment confirmation risks the inclusion of false-positive cases. When defining VTE using the NPR, anticoagulant prescription information should therefore be considered particularly for cases recorded in an outpatient setting.

Non-small cell lung cancer in young adults: presentation and survival in the English National Lung Cancer Audit

BACKGROUND

Non-small cell lung cancer (NSCLC) in young adults is a rare but devastating illness with significant socioeconomic implications, and studies of this patient subgroup are limited.

AIM

This study employed the National Lung Cancer Audit to compare the clinical features and survival of young adults with NSCLC with the older age groups.

DESIGN

A retrospective cohort review using a validated national audit dataset.

METHODS

Data were analysed for the period between 1 January 2004 and 31 December 2011. Young adults were defined as between 18 and 39 years, and all others were divided into decade age groups, up to the 80 years and above group. We performed logistic and Cox regression analyses to assess clinical outcomes.

RESULTS

Of a total of 1 46 422 patients, 651 (0.5%) were young adults, of whom a higher proportion had adenocarcinoma (48%) than in any other age group. Stage distribution of NSCLC was similar across the age groups and 71% of young patients had stage IIIb/IV. Performance status (PS) was 0-1 for 85%. Young adults were more likely to have surgery and chemotherapy compared with the older age groups and had better overall and post-operative survival. The proportion with adenocarcinoma, better PS and that receiving surgery or chemotherapy diminished progressively with advancing decade age groups.

CONCLUSION

In our cohort of young adults with NSCLC, the majority had good PS despite the same late-stage disease as older patients. They were more likely to have treatment and survive longer than older patients.

Decreased fertility rates in 9639 women diagnosed with inflammatory bowel disease: a United Kingdom population-based cohort study

BACKGROUND

Clinical studies have reported reduced fertility in women with inflammatory bowel disease (IBD).

AIM

To compare fertility rates in women with IBD to those in women without IBD and assess whether the relative fertility differed following IBD diagnosis, flares and surgery.

METHODS

Women aged 15-44 years in 1990-2010 were identified from a UK primary care database. We estimated overall and age-specific fertility rates by 5-year age bands for women with and without IBD. We used Poisson regression to calculate adjusted fertility rate ratios (AFRR), adjusted for age, smoking and socioeconomic deprivation.

RESULTS

There were 46.2 live births per 1000 person-years [95% confidence interval (95% CI); 44.6-47.9] in 9639 women with IBD and 49.3 (95% CI 49.2-49.5) in 2 131 864 without (AFRR: 0.93; 95% CI: 0.89-0.96). Excluding periods of contraception use, the AFRR was 0.99 (95% CI: 0.95-1.03). Before diagnosis, the AFRR for women with ulcerative colitis (UC) was 1.07 (95% CI: 0.99-1.16) and was 0.88 (95% CI: 0.81-0.97) for women with CD. After diagnosis, AFRRs were 0.87 (95% CI: 0.82-0.94) for CD and 0.92 (95% CI: 0.86-1.00) for UC. The fertility rate was lower following flares (AFRR: 0.70; 95% CI: 0.59-0.82) or surgery (AFRR: 0.84; 95% CI: 0.77-0.92). Women with pouch and non-pouch surgery had similar overall fertility though the reduction after surgery was greater for pouches (AFRR: 0.48; 95% CI: 0.23-0.99).

CONCLUSIONS

Women with Crohn's disease have marginally lower fertility rates. These rates decreased following flares and surgical interventions. Fertility rates returned almost to normal when women were not prescribed contraception but the reduction following surgical intervention remained. As the lifetime effect of pouch vs. nonpouch surgery on fertility is small, the reduction post-pouch surgery should be interpreted with caution.