Validation of algorithms to ascertain clinical conditions and medical procedures used during pregnancy

Postpartum Depression in Reproductive-Age Women With and Without Rheumatic Disease: A Population-Based Matched Cohort Study

OBJECTIVE

To examine postpartum depression (PPD) among women with axial spondyloarthritis (axSpA), psoriatic arthritis (PsA), or rheumatoid arthritis (RA) in comparison with a matched population without rheumatic disease (RD).

METHODS

A retrospective analysis using the 2013-2018 IBM MarketScan Commercial Claims and Encounters Database was conducted. Pregnant women with axSpA, PsA, or RA were identified, and the delivery date was used as the index date. We restricted the sample to women ≤ 55 years with continuous enrollment ≥ 6 months before date of last menstrual period and throughout pregnancy. Each patient was matched with 4 individuals without RD on: (1) maternal age at delivery, (2) prior history of depression, and (3) duration of depression before delivery. Cox frailty proportional hazards models estimated the crude and adjusted hazard ratios (aHR) and 95% CI of incident postpartum depression within 1 year among women with axSpA, PsA, or RA (axSpA/PsA/RA cohort) compared to the matched non-RD comparison group.

RESULTS

Overall, 2667 women with axSpA, PsA, or RA and 10,668 patients without any RD were included. The median follow-up time in days was 256 (IQR 93-366) and 265 (IQR 99-366) for the axSpA/PsA/RA cohort and matched non-RD comparison group, respectively. Development of PPD was more common in the axSpA/PsA/RA cohort relative to the matched non-RD comparison group (axSpA/PsA/RA cohort: 17.2%; matched non-RD comparison group: 12.8%; aHR 1.22, 95% CI 1.09-1.36).

CONCLUSION

Postpartum depression is significantly higher in women of reproductive age with axSpA/PsA/RA when compared to those without RD.

Antipsychotic Use in Early Pregnancy and the Risk of Maternal and Neonatal Complications

OBJECTIVE

To assess the association between antipsychotic use in early pregnancy and the risk of maternal and neonatal metabolic complications.

METHODS

We conducted a population-based retrospective cohort study (January 1, 2010, to December 31, 2016) using the Health and Welfare Database in Taiwan. Pregnant women (18 to 49 years of age) were grouped as antipsychotic users (ie, received oral antipsychotic monotherapy during the first 20 weeks of pregnancy) and nonusers. Antipsychotic users were further categorized into first-generation antipsychotic and second-generation antipsychotic users. Propensity score methods, including matching and inverse probability of treatment weighting, were used to balance covariates. Conditional logistic regression and Cox proportional hazards models were used to compare risks of maternal (gestational diabetes mellitus, preterm birth) and neonatal (low birth weight [LBW], macrosomia) outcomes.

RESULTS

Antipsychotic users had a notably higher risk of preterm birth compared with nonusers (adjusted HR, 1.29; 95% CI, 1.04 to 1.60), but the risk of gestational diabetes mellitus (HR, 1.21; 95% CI, 0.94 to 1.56), LBW (odds ratio [OR], 1.07; 95% CI, 0.84 to 1.37), and macrosomia (OR, 1.36; 95% CI, 0.63 to 2.92) did not differ between the two groups. Among women who received antipsychotics, the odds of LBW were significantly higher in second-generation antipsychotic users compared with first-generation antipsychotic users (adjusted OR, 1.32; 95% CI, 1.04 to 1.68).

CONCLUSION

This study found that using antipsychotics in early pregnancy did not result in a greater risk of metabolic complications both for mothers and newborns. For women requiring treatment with antipsychotics during pregnancy, they should be monitored for the risk of preterm birth and low infant birth weight.

Changing patterns in medication prescription for gestational diabetes during a time of guideline change in the USA: a cross-sectional study

OBJECTIVE

To define patterns of prescription and factors associated with choice of pharmacotherapy for gestational diabetes mellitus (GDM), namely metformin, glyburide and insulin, during a period of evolving professional guidelines.

DESING

Cross-sectional study.

SETTING

US commercial insurance beneficiaries from Market-Scan (late 2015 to 2018).

STUDY DESIGN

We included women with GDM, singleton gestations, 15-51 years of age on pharmacotherapy. The exposure was pharmacy claims for metformin, glyburide and insulin.

MAIN OUTCOMES

Pharmacotherapy for GDM with either oral agent, metformin or glyburide, compared with insulin as the reference, and secondarily, consequent treatment modification (addition and/or change) to metformin, glyburide or insulin.

RESULTS

Among 37 762 women with GDM, we analysed data from 10 407 (28%) with pharmacotherapy, 21% with metformin (n = 2147), 48% with glyburide (n = 4984) and 31% with insulin (n = 3276). From late 2015 to 2018, metformin use increased from 17 to 29%, as did insulin use from 26 to 44%, whereas glyburide use decreased from 58 to 27%. By 2018, insulin was the most common pharmacotherapy for GDM; metformin was more likely to be prescribed by 9% compared with late 2015/16, but glyburide was less likely by 45%. Treatment modification occurred in 20% of women prescribed metformin compared with 2% with insulin and 8% with glyburide.

CONCLUSIONS

Insulin followed by metformin has replaced glyburide as the most common pharmacotherapy for GDM among a privately insured US population during a time of evolving professional guidelines. Further evaluation of the relative effectiveness and safety of metformin compared with insulin is needed.

TWEETABLE ABSTRACT

Insulin followed by metformin has replaced glyburide as the most common pharmacotherapy for gestational diabetes mellitus in the USA.

Validation of a Claims-based Algorithm to Identify Pregestational Diabetes Among Pregnant Women in the United States

BACKGROUND

Identifying pregestational diabetes in pregnant women using administrative claims databases is important for studies of the safety of antidiabetic treatment in pregnancy, but limited data are available on the validity of case-identifying algorithms. The purpose of this study was to evaluate the validity of an administrative claims-based algorithm to identify pregestational diabetes.

METHODS

Using a cohort of pregnant women nested within the Medicaid Analytic Extract (MAX) database, we developed an algorithm to identify pregestational type 1 and type 2 diabetes, distinct from gestational diabetes. Within a single large healthcare system in the Boston area, we identified women who delivered an infant between 2000 and 2010 and were covered by Medicaid, and linked their electronic health records to their Medicaid claims within MAX. Medical records were reviewed by two physicians blinded to the algorithm classification to confirm or rule out pregestational diabetes, with disagreements resolved by discussion. We calculated positive predictive values with 95% confidence intervals using the medical record as the reference standard.

RESULTS

We identified 49 pregnancies classified by the claims-based algorithm as pregestational diabetes that were linked to the electronic health records and had records available for review. The PPV for any pregestational diabetes was 92% [95% confidence interval (CI) 82%, 97%], type 2 diabetes 87% (68%, 95%), and type 1 diabetes 57% (37%, 75%).

CONCLUSIONS

The claims-based algorithm for pregestational diabetes and type 2 diabetes performed well; however, the PPV was low for type 1 diabetes.

Prescription medication use and baseline health status of women with live-birth deliveries in a national data network

BACKGROUND

The US Food and Drug Administration increasingly uses administrative databases to conduct surveillance of medications used during pregnancy. To assess adverse fetal effects, administrative records must be linked between the mother and infant. The Sentinel Initiative's Mother-Infant Linkage Table provides a large cohort of linked deliveries from national, regional, and public insurance claims in the United States for the US Food and Drug Administration to conduct surveillance.

OBJECTIVE

This study aimed to describe baseline health conditions and prescription medication use during pregnancy in cohorts of women with a live-birth delivery linked and not linked to an infant.

STUDY DESIGN

Live-birth deliveries in women aged 10 to 54 years with at least 391 days of medical and drug coverage before delivery were identified in the Sentinel Mother-Infant Linkage Table from 2000 to 2019. Two cohorts were created for analysis: deliveries linked to infant records (linked deliveries, n=2,320,805) and deliveries unable to be linked to an infant (not-linked deliveries, n=504,785). Baseline health conditions, pregnancy history, healthcare utilization, and pregnancy conditions were defined using International Classification of Diseases, Ninth Revision, and International Classification of Diseases, Tenth Revision, diagnosis and procedure codes. Medication exposure was identified in a 90-day prepregnancy period and in each trimester.

RESULTS

Few notable differences were observed between the linked and not-linked deliveries except for maternal age and preterm birth; the not-linked cohort was 3.4 years younger on average and more likely to have a preterm delivery. At baseline among the linked deliveries, the most common conditions were depression and anxiety (5.2% each), acquired hypothyroidism (5.0%), pain conditions (3.9%), and asthma (2.8%). Among linked deliveries, 6.9% had evidence of gestational diabetes mellitus, 3.9% had gestational hypertension, and 4.5% had preeclampsia or eclampsia. The most commonly used medications during pregnancy in the linked deliveries were antibacterials (41.6%) and antiemetics (21.5%); similar medication use patterns were observed in the not-linked cohort. Age trends were observed for some medication groups; anti-infectives, pain medications, and antiemetics were more common in younger mothers, whereas endocrine medications were more common in older mothers.

CONCLUSION

Similarities between the linked and not-linked cohorts suggested that the ability to link mother and infant records is not influenced by maternal health status. In the linked cohort, the prevalence of specific pregnancy complications and medication use were similar to previously reported national estimates. Some baseline comorbidities, such as obesity and smoking, may be underestimated in the Sentinel Mother-Infant Linkage.

Validation of ICD-10 Codes for Gestational and Pregestational Diabetes During Pregnancy in a Large, Public Hospital

BACKGROUND

The use of billing codes (ICD-10) to identify and track cases of gestational and pregestational diabetes during pregnancy is common in clinical quality improvement, research, and surveillance. However, specific diagnoses may be misclassified using ICD-10 codes, potentially biasing estimates. The goal of this study is to provide estimates of validation parameters (sensitivity, specificity, positive predictive value, and negative predictive value) for pregestational and gestational diabetes diagnosis using ICD-10 diagnosis codes compared with medical record abstraction at a large public hospital in Atlanta, Georgia.

METHODS

This study includes 3,654 deliveries to Emory physicians at Grady Memorial Hospital in Atlanta, Georgia, between 2016 and 2018. We linked information abstracted from the medical record to ICD-10 diagnosis codes for gestational and pregestational diabetes during the delivery hospitalization. Using the medical record as the gold standard, we calculated sensitivity, specificity, positive predictive value, and negative predictive value for each.

RESULTS

For both pregestational and gestational diabetes, ICD-10 codes had a high-negative predictive value (>99%, Table 3) and specificity (>99%). For pregestational diabetes, the sensitivity was 85.9% (95% CI = 78.8, 93.0) and positive predictive value 90.8% (95% CI = 85, 97). For gestational diabetes, the sensitivity was 95% (95% CI = 92, 98) and positive predictive value 86% (95% CI = 81, 90).

CONCLUSIONS

In a large public hospital, ICD-10 codes accurately identified cases of pregestational and gestational diabetes with low numbers of false positives.

Association of Breast Implants with Nonspecific Symptoms, Connective Tissue Diseases, and Allergic Reactions: A Retrospective Cohort Analysis

BACKGROUND

Given the rising media attention regarding various adverse conditions attributed to breast implants, the authors examined the association between breast implantation and the risk of being diagnosed with connective tissue diseases, allergic reactions, and nonspecific constitutional complaints in a cohort study with longitudinal follow-up.

METHODS

Women enrolled in a regional military health care system between 2003 and 2012 were evaluated in this retrospective cohort study. A propensity score was generated to match women who underwent breast implantation with women who did not undergo breast implantation. The propensity score included age, social history, health care use, comorbidities, and medication use. Outcomes assessed included International Classification of Diseases, Ninth Revision, diagnoses codes for (1) nonspecific constitutional symptoms, (2) nonspecific cardiac conditions, (3) rheumatoid arthritis and systemic lupus erythematosus, (4) other connective tissue diseases, and (5) allergic reactions.

RESULTS

Of 22,063 women included in the study (513 breast implants and 21,550 controls), we propensity score-matched 452 breast implant recipients with 452 nonrecipients. Odds ratios and 95 percent confidence intervals in breast implant recipients compared to nonrecipients were similar, including nonspecific constitutional symptoms (OR, 0.77; 95 percent CI, 0.53 to 1.13), nonspecific cardiac conditions (OR, 0.97; 95 percent CI, 0.69 to 1.37), rheumatoid arthritis and systemic lupus erythematosus (OR, 0.66; 95 percent CI, 0.33 to 1.31), other connective tissue diseases (OR, 1.02; 95 percent CI, 0.78 to 1.32), and allergic reactions (OR, 1.18; 95 percent CI, 0.84 to 1.66).

CONCLUSIONS

Women with breast implants did not have an increased likelihood of being diagnosed with nonspecific constitutional symptoms, connective tissue disorders, and/or allergic reaction conditions.

CLINICAL QUESTION/LEVEL OF EVIDENCE

Therapeutic, III.

Factors Associated With Postpartum Diabetes Screening in Women With Gestational Diabetes and Medicaid During Pregnancy

INTRODUCTION

Women with gestational diabetes are 7 times more likely to develop type 2 diabetes and require lifelong diabetes screening. Loss of health coverage after pregnancy, as occurs in states that did not expand Medicaid, limits access to guideline-driven follow-up care and fosters health inequity. This study aims to understand the factors associated with the receipt of postpartum diabetes screening for women with gestational diabetes in a state without Medicaid expansion.

METHODS

Electronic health record and Medicaid claims data were linked to generate a retrospective cohort of 1,078 women with gestational diabetes receiving care in Federally Qualified Health Centers in Missouri from 2010 to 2015. In 2019-2020, data were analyzed to determine the factors associated with the receipt of recommended postpartum diabetes screening (fasting plasma glucose, 2-hour oral glucose tolerance test, or HbA1c in specified timeframes) using a Cox proportional hazards model through 18 months of follow-up.

RESULTS

Median age in this predominantly urban population was 28 (IQR=24-33) years. Self-reported racial or ethnic minorities comprised more than half of the population. Only 9.7% of women were screened at 12 weeks, and 20.8% were screened at 18 months. Prenatal certified diabetes education (adjusted hazard ratio=1.74, 95% CI=1.22, 2.49) and access to public transportation (adjusted hazard ratio=1.70, 95% CI=1.13, 2.54) were associated with increased screening in a model adjusted for race/ethnicity, the total number of prenatal visits, the use of diabetes medication during pregnancy, and a pregnancy-specific comorbidity index that incorporated age.

CONCLUSIONS

This study underscores the importance of access to public transportation, prenatal diabetes education, and continued healthcare coverage for women on Medicaid to support the receipt of guideline-recommended follow-up care and improve health equity.

Validated methods for identifying individuals with obesity in health care administrative databases: A systematic review

BACKGROUND

Health care administrative databases are increasingly used for health studies and public health surveillance. Cases of individuals with obesity are selected using case-identification methods. However, the validity of these methods is fragmentary and particularly challenging for obesity case identification.

OBJECTIVE

The objectives of this systematic review are to (1) determine the case-identification methods used to identify individuals with obesity in health care administrative databases and (2) to summarize the validity of these case-identification methods when compared with a reference standard.

METHODS

A systematic literature search was conducted in six bibliographic databases for the period January 1980 to June 2019 for all studies evaluating obesity case-identification methods compared with a reference standard.

RESULTS

Seventeen articles met the inclusion criteria. International Classification of Diseases (ICD) codes were the only case-identification method utilized in selected articles. The performance of obesity-identification methods varied widely across studies, with positive predictive value ranging from 19% to 100% while sensitivity ranged from 3% to 92%. The sensitivity of these methods was usually low while the specificity was higher.

CONCLUSION

When obesity is reported in health care administrative databases, it is usually correctly reported; however, obesity tends to be highly underreported in databases. Therefore, case-identification methods to monitor the prevalence and incidence of obesity within health care administrative databases are not reliable. In contrast, the use of these methods remains relevant for the selection of individuals with obesity for cohort studies, particularly when identifying cohorts of individuals with severe obesity or cohorts where obesity is associated with comorbidities.

Identifying pregnancies in insurance claims data: Methods and application to retinoid teratogenic surveillance

PURPOSE

The purpose of the study is to develop an algorithm to identify pregnancies in administrative databases and apply it to assess pregnancy rates and outcomes in women prescribed isotretinoin or tretinoin.

METHODS

Using the 2011 to 2015 Truven Health MarketScan Database, we identified pregnancies, including losses and terminations. In a cohort design, nonpregnant women filling a prescription for isotretinoin or tretinoin were matched to five women without either prescription. Women were followed for 365 days or until conception, medication discontinuation, or enrollment discontinuation ("prescription episode"). Rates of pregnancy, risks of pregnancy losses, and prevalence of infant malformations at birth were assessed by exposure.

RESULTS

We identified 2 179 192 livebirths, 8434 stillbirths, 2521 mixed births, 415 110 spontaneous abortions, 124 556 elective terminations, and 8974 unspecified abortions. There were 86 834 isotretinoin and 973 587 tretinoin episodes, matched to 5 302 105 unexposed women. Pregnancy rates were 3 (isotretinoin), 19 (tretinoin), and 34 (unexposed) per 1000 person-years. Risk of spontaneous pregnancy losses were similar; however, terminations were more common in the isotretinoin-exposed (28% [95% CI: 21%-36%]) than the tretinoin-exposed (10% [95% CI: 9%-11%]) or unexposed pregnancies (6%). Malformations occurred in 4.5% (95% CI: 3.5%-5.6%) of the tretinoin-exposed pregnancies and 4.2% of the unexposed pregnancies (adjusted odds ratio: 1.16 [95% CI: 0.85-1.58]); isotretinoin-exposed births were too few to assess malformations.

CONCLUSIONS

Administrative databases can complement risk evaluation and mitigation strategies (REMS) for known teratogens and contribute to safety surveillance for other medications. Here, isotretinoin-exposed pregnancy rates were low, but existent, and many pregnancies were terminated. Tretinoin exposure was not associated with a meaningfully elevated risk of losses or malformations as compared with unexposed pregnancies.

Postpartum diabetes screening among low income women with gestational diabetes in Missouri 2010-2015

Background

Gestational diabetes increases risk for type 2 diabetes seven-fold, creating a large public health burden in a young population. In the US, there are no large registries for tracking postpartum diabetes screening among women in under-resourced communities who face challenges with access to care after pregnancy. Existing data from Medicaid claims is limited as women often lose this coverage within months of delivery. In this study, we aim to leverage data from electronic health records and administrative claims to better assess postpartum diabetes screening rates among low income women.

Methods

A retrospective population of 1078 women with gestational diabetes who delivered between 1/1/2010 and 10/8/2015 was generated by linking electronic health record data from 21 Missouri Federally Qualified Health Centers (FQHCs) with Medicaid administrative claims. Screening rates for diabetes were calculated within 12 weeks and 1 year of delivery. Initial screening after the first postpartum year was also documented.

Results

Median age in the final population was 28 (IQR 24–33) years with over-representation of black non-Hispanic and urban women. In the final population, 9.7% of women had a recommended diabetes screening test within 12 weeks and 18.9% were screened within 1 year of delivery. An additional 125 women received recommended screening for the first time beyond 1 year postpartum. The percentage of women who had a postpartum visit (83.9%) and any glucose testing (40.6%) in the first year far exceeded the proportion of women with recommended screening tests.

Conclusions

Linking electronic health record and administrative claims data provides a more complete picture of healthcare follow-up among low income women after gestational diabetes. While screening rates are higher than reported with claims data alone, there are opportunities to improve adherence to screening guidelines in this population.

Electronic supplementary material

The online version of this article (10.1186/s12889-019-6475-0) contains supplementary material, which is available to authorized users.

Validation of data from electronic data warehouse in diabetic ketoacidosis: Caution is needed

AIMS

This study validated enterprise data warehouse (EDW) data for a cohort of hospitalized patients with a primary diagnosis of diabetic ketoacidosis (DKA).

METHODS

247 patients with 319 admissions for DKA (ICD-9 code 250.12, 250.13, or 250.xx with biochemical criteria for DKA) were admitted to Northwestern Memorial Hospital from 1/1/2010 to 9/1/2013. Validation was performed by electronic medical record (EMR) review of 10% of admissions (N = 32). Classification of diabetes type (Type 1 vs. Type 2) and DKA clinical status were compared between the EMR review and EDW data.

RESULTS

Key findings included incorrect classification of diabetes type in 5 of 32 (16%) admissions and indeterminable classification in 5 admissions. DKA was not present, based on the review, in 11 of 32 (34%) admissions. DKA was not present, based on biochemical criteria, in 15 of 32 (47%) admissions.

CONCLUSIONS

This study found that EDW data have substantial errors. Some discrepancies can be addressed by refining the EDW query code, while others, related to diabetes classification and DKA diagnosis, cannot be corrected without improving clinical coding accuracy, consistency of medical record documentation, or EMR design. These results support the need for comprehensive validation of data for complex clinical populations obtained through data repositories such as the EDW.

Continuation of Atypical Antipsychotic Medication During Early Pregnancy and the Risk of Gestational Diabetes

OBJECTIVE

Some atypical antipsychotics are associated with metabolic side effects, which are risk factors for gestational diabetes. The authors examined the risk of developing gestational diabetes associated with the continuation of treatment with aripiprazole, ziprasidone, quetiapine, risperidone, and olanzapine during pregnancy compared with discontinuation of these antipsychotic drugs.

METHOD

Nondiabetic pregnant women who were linked to a live-born infant and enrolled in Medicaid (2000-2010) and who received one or more prescriptions dispensed for an antipsychotic drug during the 3 months before pregnancy were included in the analyses. Among 1,543,334 pregnancies, some expectant mothers at baseline were receiving treatment with aripiprazole (N=1,924), ziprasidone (N=673), quetiapine (N=4,533), risperidone (N=1,824), or olanzapine (N=1,425). For each antipsychotic drug, women with two or more dispensings ("continuers") were compared with women with no dispensings ("discontinuers") during the first half of pregnancy. A generalized linear model and propensity-score stratification were used to obtain absolute and relative risks of developing gestational diabetes, with adjustment for confounders.

RESULTS

Women who continued antipsychotic treatment during pregnancy generally had higher comorbidity and longer baseline antipsychotic use. The crude risk of developing gestational diabetes among continuers compared with discontinuers, respectively, was 4.8% and 4.5% for aripiprazole, 4.2% and 3.8% for ziprasidone, 7.1% and 4.1% for quetiapine, 6.4% and 4.1% for risperidone, and 12.0% and 4.7% for olanzapine. The adjusted relative risks were 0.82 (95% CI=0.50-1.33) for aripiprazole, 0.76 (95% CI=0.29-2.00) for ziprasidone, 1.28 (95% CI=1.01-1.62) for quetiapine, 1.09 (95% CI=0.70-1.70) for risperidone, and 1.61 (95% CI=1.13-2.29) for olanzapine.

CONCLUSIONS

Compared with women who discontinued use of an atypical antipsychotic medication before the start of pregnancy, women who continued treatment with olanzapine or quetiapine had an increased risk of gestational diabetes that may be explained by the metabolic effects associated with these two drugs.

Association between Off-Peak Hour Birth and Neonatal Morbidity and Mortality among Very Low Birth Weight Infants

OBJECTIVE

To assess the independent association between overnight or "off-peak" hour delivery and 3 neonatal morbidities strongly associated with childhood neurocognitive impairment.

STUDY DESIGN

Retrospective population based cohort study of all infants with birth weights of 500-1499 g born without severe congenital anomalies in California or Pennsylvania between 2002 and 2009. Off-peak hour delivery was defined as birth between 12:00 a.m. and 6:59 a.m. The study outcomes were death; bronchopulmonary dysplasia, retinopathy of prematurity, and severe (grade 3 or 4) intraventricular hemorrhage among survivors; the composite of each morbidity or mortality; and the composite of death or 1 or more of the evaluated morbidities.

RESULTS

Of 47 617 evaluated infants, 9317 (19.6%) were born during off-peak hours. The frequencies of all study outcomes were higher among infants born during off-peak compared with peak hours. After adjusting for maternal, infant, and hospital characteristics, off-peak hour delivery was associated with increased odds of severe intraventricular hemorrhage among survivors (OR 1.39, 95% CI 1.23-1.57) and the composite outcomes of death or severe intraventricular hemorrhage (OR 1.16, 95% CI 1.07-1.25) and death or major morbidity (OR 1.08, 95% CI 1.02-1.15). There was no evidence of subgroup effects based on delivery mode, birth hospital neonatal intensive care level or annual very low birth weight infant delivery volume, or weekday vs weekend off-peak hour delivery for any study outcome.

CONCLUSIONS

Very low birth weight infants born between midnight and 7:00 a.m. are at increased risk for severe intraventricular hemorrhage and death or major neonatal morbidity.

Validation of administrative and clinical case definitions for gestational diabetes mellitus against laboratory results

AIM

To examine the validity of International Classification of Disease, version 10 (ICD-10) codes for gestational diabetes mellitus in administrative databases (outpatient and inpatient), and in a clinical perinatal database (Alberta Perinatal Health Program), using laboratory data as the 'gold standard'.

METHODS

Women aged 12-54 years with in-hospital, singleton deliveries between 1 October 2008 and 31 March 2010 in Alberta, Canada were included in the study. A gestational diabetes diagnosis was defined in the laboratory data as ≥2 abnormal values on a 75-g oral glucose tolerance test or a 50-g glucose screen ≥10.3 mmol/l.

RESULTS

Of 58 338 pregnancies, 2085 (3.6%) met gestational diabetes criteria based on laboratory data. The gestational diabetes rates in outpatient only, inpatient only, outpatient or inpatient combined, and Alberta Perinatal Health Program databases were 5.2% (3051), 4.8% (2791), 5.8% (3367) and 4.8% (2825), respectively. Although the outpatient or inpatient combined data achieved the highest sensitivity (92%) and specificity (97%), it was associated with a positive predictive value of only 57%. The majority of the false-positives (78%), however, had one abnormal value on oral glucose tolerance test, corresponding to a diagnosis of impaired glucose tolerance in pregnancy.

CONCLUSIONS

The ICD-10 codes for gestational diabetes in administrative databases, especially when outpatient and inpatient databases are combined, can be used to reliably estimate the burden of the disease at the population level. Because impaired glucose tolerance in pregnancy and gestational diabetes may be managed similarly in clinical practice, impaired glucose tolerance in pregnancy is often coded as gestational diabetes.

Distributed data networks: a blueprint for Big Data sharing and healthcare analytics

This paper defines the attributes of distributed data networks and outlines the data and analytic infrastructure needed to build and maintain a successful network. We use examples from one successful implementation of a large-scale, multisite, healthcare-related distributed data network, the U.S. Food and Drug Administration-sponsored Sentinel Initiative. Analytic infrastructure-development concepts are discussed from the perspective of promoting six pillars of analytic infrastructure: consistency, reusability, flexibility, scalability, transparency, and reproducibility. This paper also introduces one use case for machine learning algorithm development to fully utilize and advance the portfolio of population health analytics, particularly those using multisite administrative data sources.

Developing an Algorithm to Identify History of Cancer Using Electronic Medical Records

Introduction/Objective:

The objective of this study was to develop an algorithm to identify Kaiser Permanente Colorado (KPCO) members with a history of cancer.

Background:

Tumor registries are used with high precision to identify incident cancer, but are not designed to capture prevalent cancer within a population. We sought to identify a cohort of adults with no history of cancer, and thus, we could not rely solely on the tumor registry.

Methods:

We included all KPCO members between the ages of 40–75 years who were continuously enrolled during 2013 (N=201,787). Data from the tumor registry, chemotherapy files, inpatient and outpatient claims were used to create an algorithm to identify members with a high likelihood of cancer. We validated the algorithm using chart review and calculated sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for occurrence of cancer.

Findings:

The final version of the algorithm achieved a sensitivity of 100 percent and specificity of 84.6 percent for identifying cancer. If we relied on the tumor registry alone, 47 percent of those with a history of cancer would have been missed.

Discussion:

Using the tumor registry alone to identify a cohort of patients with prior cancer is not sufficient. In the final version of the algorithm, the sensitivity and PPV were improved when a diagnosis code for cancer was required to accompany oncology visits or chemotherapy administration.

Conclusion:

Electronic medical record (EMR) data can be used effectively in combination with data from the tumor registry to identify health plan members with a history of cancer.

Obesity is independently associated with infection in hospitalised patients with end-stage liver disease

BACKGROUND

Infection is the most common cause of mortality in end-stage liver disease (ESLD). The impact of obesity on infection risk in ESLD is not established.

AIM

To characterise the impact of obesity on infection risk in ESLD.

METHODS

We evaluated the association between infection and obesity in patients with ESLD. Patients grouped as non-obese, obesity class I-II and obesity class III were studied using the Nationwide Inpatient Sample. Validated diagnostic code based algorithms were utilised to determine weight category and infections, including bacteraemia, skin/soft tissue infection, urinary tract infection (UTI), pneumonia/respiratory infection, Clostridium difficile infection (CDI) and spontaneous bacterial peritonitis (SBP). Risk factors for infection and mortality were assessed using multivariable logistic regression analysis.

RESULTS

Of 115 465 patients identified, 100 957 (87.5%) were non-obese and 14 508 (12.5%) were obese, with 9489 (8.2%) as obesity class I-II and 5019 (4.3%) as obesity class III. 37 117 patients (32.1%) had an infection diagnosis. Infection was most prevalent among obesity class III (44.0%), followed by obesity class I-II (38.9%) and then non-obese (31.9%). In multivariable modelling, class III obesity (OR = 1.41; 95% CI 1.32-1.51; P < 0.001), and class I-II obesity (OR = 1.08; 95% CI 1.01-1.15; P = 0.026) were associated with infection. Compared to non-obese patients, obese individuals had greater prevalence of bacteraemia, UTI, and skin/soft tissue infection as compared to non-obese patients.

CONCLUSIONS

Obesity is newly identified to be independently associated with infection in end-stage liver disease. The distribution of infection sites varies based on weight category.

Validation of an obstetric comorbidity index in an external population

Objectives

An obstetric comorbidity index has been developed recently with superior performance characteristics relative to general comorbidity measures in an obstetric population. This study aimed to externally validate this index and to examine the impact of including hospitalisation/delivery records only when estimating comorbidity prevalence and discriminative performance of the obstetric comorbidity index.

Design

Validation study.

Setting

Alberta, Canada.

Population

Pregnant women who delivered a live or stillborn infant in hospital (n = 5995).

Methods

Administrative databases were linked to create a population‐based cohort. Comorbid conditions were identified from diagnoses for the delivery hospitalisation, all hospitalisations and all healthcare contacts (i.e. hospitalisations, emergency room visits and physician visits) that occurred during pregnancy and 3 months pre‐conception. Logistic regression was used to test the discriminative performance of the comorbidity index.

Main outcome measures

Maternal end‐organ damage and extended length of stay for delivery.

Results

Although prevalence estimates for comorbid conditions were consistently lower in delivery records and hospitalisation data than in data for all healthcare contacts, the discriminative performance of the comorbidity index was constant for maternal end‐organ damage [all healthcare contacts area under the receiver operating characteristic curve (AUC) = 0.70; hospitalisation data AUC = 0.67; delivery data AUC = 0.65] and extended length of stay for delivery (all healthcare contacts AUC = 0.60; hospitalisation data AUC = 0.58; delivery data AUC = 0.58).

Conclusions

The obstetric comorbidity index shows similar performance characteristics in an external population and is a valid measure of comorbidity in an obstetric population. Furthermore, the discriminative performance of the comorbidity index was similar for comorbidities ascertained at the time of delivery, in hospitalisation data or through all healthcare contacts.

Trends in glyburide compared with insulin use for gestational diabetes treatment in the United States, 2000-2011

OBJECTIVE

To describe trends and identify factors associated with choice of pharmacotherapy for gestational diabetes (GDM) from 2000-2011 using a healthcare claims database.

METHODS

This was a retrospective cohort study of a large nationwide population of commercially insured women with GDM and pharmacy claims for glyburide or insulin before delivery, 2000-2011. We excluded women younger than 15 years or older than 50 years, those with prior noninsulin-dependent diabetes mellitus, or those who had multiple gestations. We estimated trends over time in the use of glyburide compared with insulin and prevalence ratios and 95% confidence intervals (CIs) for the association between covariates of interest and treatment with glyburide compared with insulin.

RESULTS

We identified 10,778 women with GDM treated with glyburide (n=5,873) or insulin (n=4,905). From 2000 to 2011, glyburide use increased from 7.4% to 64.5%, becoming the more common treatment in 2007. Women less likely to be treated with glyburide were those with metabolic syndrome (prevalence ratio 0.71, 95% CI 0.50-0.99), hyperandrogenism (prevalence ratio 0.77, 95% CI 0.62-0.97), polycystic ovarian syndrome (prevalence ratio 0.88, 95% CI 0.78-0.99), hypothyroidism (prevalence ratio 0.89, 95% CI 0.83-0.96), or undergoing infertility treatment (prevalence ratio 0.93, 95% CI 0.86-1.02). The probability of receiving glyburide decreased by 5% for every 10-year increase in maternal age (prevalence ratio 0.95, 95% CI 0.91-0.99). Among women prescribed with glyburide, 7.8% switched or augmented to a different drug class compared with 1.1% of insulin initiators.

CONCLUSION

Glyburide has replaced insulin as the more common pharmacotherapy for GDM over the past decade among those privately insured. Given its rapid uptake and the potential implications of suboptimal glucose control on maternal and neonatal health, robust evaluation of glyburide's relative effectiveness is warranted to inform treatment decisions for women with gestational diabetes.

LEVEL OF EVIDENCE

II.

The association between parity and birthweight in a longitudinal consecutive pregnancy cohort

BACKGROUND

Nulliparity is associated with lower birthweight, but few studies have examined how within-mother changes in risk factors impact this association.

METHODS

We used longitudinal electronic medical record data from a hospital-based cohort of consecutive singleton live births from 2002-2010 in Utah. To reduce bias from unobserved pregnancies, primary analyses were limited to 9484 women who entered nulliparous from 2002-2004, with 23,380 pregnancies up to parity 3. Unrestricted secondary analyses used 101,225 pregnancies from 45,212 women with pregnancies up to parity 7. We calculated gestational age and sex-specific birthweight z-scores with nulliparas as the reference. Using linear mixed models, we estimated birthweight z-score by parity adjusting for pregnancy-specific sociodemographics, smoking, alcohol, prepregnancy body mass index, gestational weight gain, and medical conditions.

RESULTS

Compared with nulliparas', infants of primiparas were larger by 0.20 unadjusted z-score units [95% confidence interval (CI) 0.18, 0.22]; the adjusted increase was similar at 0.18 z-score units [95% CI 0.15, 0.20]. Birthweight continued to increase up to parity 3, but with a smaller difference (parity 3 vs. 0 β = 0.27 [95% CI 0.20, 0.34]). In the unrestricted secondary sample, there was significant departure in linearity from parity 1 to 7 (P < 0.001); birthweight increased only up to parity 4 (parity 4 vs. 0 β = 0.34 [95% CI 0.31, 0.37]).

CONCLUSIONS

The association between parity and birthweight was non-linear with the greatest increase observed between first- and second-born infants of the same mother. Adjustment for changes in weight or chronic diseases did not change the relationship between parity and birthweight.

Patterns of obstetric infection rates in a large sample of US hospitals

OBJECTIVE

Maternal infection is a common complication of childbirth, yet little is known about the extent to which infection rates vary among hospitals. We estimated hospital-level risk-adjusted maternal infection rates (RAIR) in a large sample of US hospitals and explored associations between RAIR and select hospital features.

STUDY DESIGN

This retrospective cohort study included hospitals in the Perspective database with >100 deliveries over 2 years. Using a composite measure of infection, we estimated and compared RAIR across hospitals using hierarchical generalized linear models. We then estimated the amount of variation in RAIR attributable to hospital features.

RESULTS

Of the 1,001,189 deliveries at 355 hospitals, 4.1% were complicated by infection. Patients aged 15-19 years were 50% more likely to experience infection than those aged 25-29 years. Rupture of membranes >24 hours (odds ratio [OR], 3.0; 95% confidence interval [CI], 3.24-3.5), unengaged fetal head (OR, 3.11; 95% CI, 2.97-3.27), and blood loss anemia (OR, 2.42; 95% CI, 2.34-2.49) had the highest OR among comorbidities commonly found in patients with infection. RAIR ranged from 1.0-14.4% (median, 4.0%; interquartile range, 2.8-5.7%). Hospital features such as geographic region, teaching status, urban setting, and higher number of obstetric beds were associated with higher infection rates, accounting for 14.8% of the variation observed.

CONCLUSION

Obstetric RAIR vary among hospitals, suggesting an opportunity to improve obstetric quality of care. Hospital features such as region, number of obstetric beds, and teaching status account for only a small portion of the observed variation in infection rates.

Prevalence, trends, and patterns of use of antidiabetic medications among pregnant women, 2001-2007

OBJECTIVE

To describe the prevalence, trends, and patterns in use of antidiabetic medications to treat hyperglycemia and insulin resistance before and during pregnancy in a large U.S. cohort of insured pregnant women.

METHODS

Pregnancies resulting in live births were identified (N=437,950) from 2001 to 2007 among 372,543 females 12-50 years of age at delivery from 10 health maintenance organizations participating in the Medication Exposure in Pregnancy Risk Evaluation Program. Information for these descriptive analyses, including all antidiabetic medications dispensed during this period, was extracted from electronic health records and newborn birth certificates.

RESULTS

A little more than 1% (1.21%) of deliveries were to women dispensed antidiabetic medication in the 120 days before pregnancy. Use of antidiabetic medications before pregnancy increased from 0.66% of deliveries in 2001 to 1.66% of deliveries in 2007 (P<.001) because of an increase in metformin use. Most women using metformin before pregnancy had a diagnosis code for polycystic ovaries or female infertility (67.2%), whereas only 13.6% had a diagnosis code for diabetes. The use of antidiabetic medications during the second or third trimester of pregnancy increased from 2.8% of deliveries in 2001 to 3.6% in 2007 (P<.001). Approximately two thirds (68%) of women using metformin before pregnancy did not use any antidiabetic medications during pregnancy.

CONCLUSIONS

Antidiabetic medication use before and during pregnancy increased from 2001 to 2007, possibly because of increasing prevalence of gestational diabetes mellitus, type 1 and type 2 diabetes, and other conditions associated with insulin resistance.

LEVEL OF EVIDENCE

III.

Extent of maternal morbidity in a managed care population in georgia

BACKGROUND

Although maternal deaths are among the most tragic events related to pregnancy, they are uncommon in the US and, therefore, inadequate indicators of a woman's pregnancy-related health. Maternal morbidity has become a more useful measure for surveillance and research. Traditional attempts to monitor maternal morbidity have used hospital discharge data, which include data only on complications that resulted in hospitalisation, underestimating the frequency and scope of complications.

METHODS

To obtain a more accurate assessment of morbidity, we applied a validated computerised algorithm to identify pregnancies and pregnancy-related complications in a defined population enrolled in a health maintenance organisation in the south-eastern US. We examined the most common morbidities by pregnancy outcome and maternal characteristics.

RESULTS

We identified 37 741 pregnancies; in half (50.7%), at least one complication occurred. The five most common were urinary tract infections, anaemia, mental health conditions, pelvic and perineal complications, and obstetrical infections. Complications were more likely in women with low socio-economic status (SES), and among non-Hispanic Black women compared with non-Hispanic White women. Multivariable models stratified by race/ethnicity indicated that in pregnancies among non-Hispanic White women, low SES had a modest effect on the odds of having preexisting medical conditions [adjusted odd ratio (AOR) 1.3 [95% confidence interval (CI) 1.2, 1.5]] or having any morbidity (AOR 1.3 [95% CI 1.2, 1.4]). Low SES had little effect on complications among non-Hispanic Black women.

CONCLUSION

Our findings suggest that comprehensive health insurance coverage may lessen the unfavourable impact of socio-economic disadvantage on the risk of maternal morbidity.