Initiation Patterns and Transitions Among Adults Using Stimulant Drugs: Latent Transition Analysis.

BACKGROUND: The fourth wave of the drug overdose epidemic in the United States includes increasing rates of stimulant-involved overdose. Recent studies of transitions leading to stimulant misuse have shown complex patterns that are not universally applicable because they have isolated individual populations or individual behaviors. A comprehensive analysis of transitions between behaviors and the associations with present-day problematic drug use has not been conducted. OBJECTIVE: This study aims to determine whether adults from the general population who use stimulants initiate use through a heterogeneous combination of behaviors and quantify the association between these typologies with present-day problematic drug use. METHODS: Individuals who have reported use of any stimulant in their lifetime were recruited from the 2021 Survey of Nonmedical Use of Prescription Drugs Program, a nationally representative web-based survey on drug use, to participate in a rapid follow-up survey about their past stimulant use. Individuals were asked which stimulants they used, the reasons for use, the routes of administration, and the sources of the stimulant. For each stimulant-related behavior, they were asked at what age, between 6 and 30 years, they initiated each behavior in a 6-year time window. A latent transition analysis was used to characterize heterogeneity in initiation typologies. Mutually exclusive pathways of initiation were identified manually by the researchers. The association of these pathways with present-day problematic drug use was calculated using logistic regression adjusted by the current age of the respondent. RESULTS: From a total of 1329 participants, 740 (55.7%) reported lifetime prescription stimulant use and 1077 (81%) reported lifetime illicit stimulant use. Three typologies were identified. The first typology was characterized by illicit stimulant initiation to get high, usually via oral or snorting routes and acquisition from friends or family or a dealer (illicit experimentation). The second typology was characterized by low, but approximately equal probabilities of initiating 1-2 new behaviors in a time window, but no singular set of behaviors characterized the typology (conservative initiation). The third was characterized by a high probability of initiating many diverse combinations of behaviors (nondiscriminatory experimentation). The choice of drug initiated was not a strong differentiator. Categorization of pathways showed those who were only in an illicit experimentation status (reference) had the lowest odds of having severe present-day problematic drug use. Odds were higher for a conservative initiation-only status (odds ratio [OR] 1.84, 95% CI 1.14-2.94), which is higher still for those moving from illicit experimentation to conservative initiation (OR 3.50, 95% CI 2.13-5.74), and highest for a nondiscriminatory experimentation status (OR 5.45, 95% CI 3.39-8.77). CONCLUSIONS: Initiation of stimulant-related use behaviors occurred across many time windows, indicating that multiple intervention opportunities are presented. Screening should be continued throughout adulthood to address unhealthy drug use before developing into full substance use disorders.

Differing Behaviors Around Adult Nonmedical Use of Prescription Stimulants and Opioids: Latent Class Analysis.

BACKGROUND: The availability of central nervous system stimulants has risen in recent years, along with increased dispensing of stimulants for treatment of, for example, parent-reported attention-deficit/hyperactivity disorder in children and new diagnoses during adulthood. Typologies of drug use, as has been done with opioids, fail to include a sufficient range of behavioral factors to contextualize person-centric circumstances surrounding drug use. Understanding these patterns across drug classes would bring public health and regulatory practices toward precision public health. OBJECTIVE: The objective of this study was to quantitatively delineate the unique behavioral profiles of adults who currently nonmedically use stimulants and opioids using a latent class analysis and to contrast the differences in findings by class. We further evaluated whether the subgroups identified were associated with an increased Drug Abuse Screening Test-10 (DAST-10) score, which is an indicator of average problematic drug use. METHODS: This study used a national cross-sectional web-based survey, using 3 survey launches from 2019 to 2020 (before the COVID-19 pandemic). Data from adults who reported nonmedical use of prescription stimulants (n=2083) or prescription opioids (n=6127) in the last 12 months were analyzed. A weighted latent class analysis was used to identify the patterns of use. Drug types, motivations, and behaviors were factors in the model, which characterized unique classes of behavior. RESULTS: Five stimulant nonmedical use classes were identified: amphetamine self-medication, network-sourced stimulant for alertness, nonamphetamine performance use, recreational use, and nondiscriminatory behaviors. The drug used nonmedically, acquisition through a friend or family member, and use to get high were strong differentiators among the stimulant classes. The latter 4 classes had significantly higher DAST-10 scores than amphetamine self-medication (P<.001). In addition, 4 opioid nonmedical use classes were identified: moderate pain with low mental health burden, high pain with higher mental health burden, risky behaviors with diverse motivations, and nondiscriminatory behaviors. There was a progressive and significant increase in DAST-10 scores across classes (P<.001). The potency of the opioid, pain history, the routes of administration, and psychoactive effect behaviors were strong differentiators among the opioid classes. CONCLUSIONS: A more precise understanding of how behaviors tend to co-occur would improve efficacy and efficiency in developing interventions and supporting the overall health of those who use drugs, and it would improve communication with, and connection to, those at risk for severe drug outcomes.

Nonmedical use of benzodiazepines and Z-drugs in the UK.

AIMS: To estimate prevalence of last 12-month nonmedical use (NMU) of benzodiazepines and Z-drugs (the nonbenzodiazepine hypnotics zaleplon, zolpidem and zopiclone) in the UK. METHODS: Data were collected using the Non-Medical Use of Prescription Drugs survey with poststratification weighting applied to be representative of the UK population (≥16 years). Participants were questioned about whether they had nonmedically used benzodiazepines and/or Z-drugs in the last 12-months and from where they had obtained the drug (including via a prescription, or illicitly from a friend/family member, a dealer or via the internet). Additional questions were asked about last 12-month use of illicit drugs (cannabis, cocaine, 3,4-methylenedioxymethylamphetamine [MDMA], non-pharmaceutical amphetamine, crack cocaine and/or heroin). RESULTS: The study included 10 006 eligible participants representing approximately 52 927 000 UK adults. The estimated prevalence of past 12-month NMU of any benzodiazepine and/or Z-drug was 1.2% (95% confidence interval: 1.0-1.5) corresponding to approximately 635 000 adults; amongst this group only an estimated 4.6% (1.2-8.0) had NMU of both a benzodiazepine and a Z-drug. The highest prevalence of NMU for only Z-drugs was among those who had used heroin in the last 12-months (5.4%, 2.7-10.5), whilst the highest prevalence of NMU for only benzodiazepines was among those who had used illicit stimulants in the last 12-months: cocaine (5.9%, 3.8-8.9), amphetamine (5.6%, 3.1-10.0) and MDMA (5.2%, 3.1-8.8). The drug non-medically used was more commonly acquired without than with a prescription for both only benzodiazepines (70.2%, 59.4-81.1 compared to 51.3%, 41.5-64.6) and only Z-drugs (75.6%, 61.6-89.7 compared to 33.9%, 16.9-51.0). CONCLUSION: There is little overlap between benzodiazepine and Z-drug NMU suggesting distinct nonmedical use of the drugs; future studies need to explore whether this relates to personal preference, drug availability or other factors. A significant proportion are acquiring these drugs for NMU without a prescription, so without guidance and monitoring from a medical practitioner. While the dangers of mixing benzodiazepines and heroin/other opioids are well documented, there is a paucity of data regarding concomitant NMU of benzodiazepines and stimulant drugs, or NMU of Z-drugs and opioids, and, given the prevalence of these combinations, this requires further investigation.

Drug product dispensing and estimates of use in a general population survey as a signal detection problem.

PURPOSE: Understanding potential bias due to rarity of the outcome is important when monitoring newly approved drugs and drugs with low availability to the general public. Although there is an increasing use of online surveys to investigate health outcomes, the limits of inference due to drug availability have not been studied. The goal of this study was to quantify the relationship between dispensing of prescription drugs and estimates of use in an online general population survey. METHODS: An online repeated, cross-sectional survey from 2018 to 2020 was used to estimate the number of adults in the United States who used prescription drugs in the general population and compared to estimated number of prescriptions dispensed over an equivalent time period. Joinpoint regression was used to quantify thresholds. A sample of respondents was retested to estimate reliability statistics. RESULTS: A model with a single threshold was the best fit, with the estimated threshold of 565 000 (95% CI: 9500-11 600 000) prescriptions dispensed per year. Above the threshold, there was a significant association between dispensing and estimates (p < 0.001); below the threshold, the relationship was not significant (p = 0.912). Above the threshold, responses were more reliable than random chance, and reliability steadily increased with increased dispensing. CONCLUSIONS: These results suggest the threshold demarcates two distinct pharmacoepidemiological paradigms when investigating drug use in general population surveys. Dispensing can be used as a guide to determine the epidemiological paradigm that is best suited.

Biochemically confirmed smoking cessation and gestational weight gain.

BACKGROUND: Prenatal smoking cessation has substantial health benefits for mothers and offspring, but concerns about weight gain may be a barrier to quitting. We quantified gestational weight gain associated with biochemically confirmed smoking cessation. METHODS: Data originated from a randomized controlled cessation trial: Smoking Cessation in Pregnancy project (1987-1991). We calculated gestational weight gain using self-reported prepregnancy weight and measured weight at 30-34 weeks of gestation. We used linear regression to estimate adjusted mean differences in gain for quitters versus continuing smokers by the last trimester. The effects of quitting earlier (by 2nd trimester) versus later (by 3rd trimester) were calculated. We assessed the percentages who gained weight according to Institute of Medicine (IOM) recommendations within 2 weeks of a full-term delivery. RESULTS: At 30-34 weeks, nulliparous and multiparous quitters gained an average of 3.0 pounds (95% CI 0.9-5.1 pounds) (1.4 kg [0.4-2.3 kg]) and 6.6 pounds (95% CI 4.3-8.9 pounds) (3.0 kg [1.9-4.0 kg]) more, respectively, than continuing smokers. Weight gain in early quitters did not differ significantly from that in late quitters. Quitters were more likely than continuing smokers to gain above current guidelines (60.3% vs 46.3%) and were less likely to gain below guidelines (11.5% vs 21.6%) (P = 0.002). CONCLUSIONS: Although quitters had modest additional weight gain by 30-34 weeks compared to continuing smokers, a high proportion in both groups gained in excess of IOM recommendations. Both quitters and continuing smokers may need support to achieve optimal gestational weight gain.

An Online Survey for Pharmacoepidemiological Investigation (Survey of Non-Medical Use of Prescription Drugs Program): Validation Study.

BACKGROUND: In rapidly changing fields such as the study of drug use, the need for accurate and timely data is paramount to properly inform policy and intervention decisions. Trends in drug use can change rapidly by month, and using study designs with flexible modules could present advantages. Timely data from online panels can inform proactive interventions against emerging trends, leading to a faster public response. However, threats to validity from using online panels must be addressed to create accurate estimates. OBJECTIVE: The objective of this study was to demonstrate a comprehensive methodological approach that optimizes a nonprobability, online opt-in sample to provide timely, accurate national estimates on prevalence of drug use. METHODS: The Survey of Non-Medical Use of Prescription Drugs Program from the Researched Abuse, Diversion and Addiction Related Surveillance (RADARS) System is an online, cross-sectional survey on drug use in the United States, and several best practices were implemented. To optimize final estimates, two best practices were investigated in detail: exclusion of respondents showing careless or improbable responding patterns and calibration of weights. The approach in this work was to cumulatively implement each method, which improved key estimates during the third quarter 2018 survey launch. Cutoffs for five exclusion criteria were tested. Using a series of benchmarks, average relative bias and changes in bias were calculated for 33 different weighting variable combinations. RESULTS: There were 148,274 invitations sent to panelists, with 40,021 who initiated the survey (26.99%). After eligibility assessment, 20.23% (29,998/148,274) of the completed questionnaires were available for analysis. A total of 0.52% (157/29,998) of respondents were excluded based on careless or improbable responses; however, these exclusions had larger impacts on lower volume drugs. Number of exclusions applied were negatively correlated to total dispensing volume by drug (Spearman ρ=-.88, P<.001). A weighting scheme including three demographic and two health characteristics reduced average relative bias by 31.2%. After weighting, estimates of drug use decreased, reflecting a weighted sample that had healthier benchmarks than the unweighted sample. CONCLUSIONS: Our study illustrates a new approach to using nonprobability online panels to achieve national prevalence estimates for drug abuse. We were able to overcome challenges with using nonprobability internet samples, including misclassification due to improbable responses. Final drug use and health estimates demonstrated concurrent validity to national probability-based drug use and health surveys. Inclusion of multiple best practices cumulatively improved the estimates generated. This method can bridge the information gap when there is a need for prompt, accurate national data.

UK survey of non-medical use of prescription drugs (NMURx) as a valuable source of general population illicit drug use data.

PURPOSE OF STUDY: The aim of the study is to describe the prevalence of illicit drug use in England and Wales using data from the UK Survey of Non-Medical Use of Prescription Drugs (NMURx) programme and to compare against the well-established Crime Survey England and Wales (CSEW). The rationale is that recreational and illicit drug use is common, but the prevalence is difficult to estimate with personal interviewing methods. STUDY DESIGN: We compared two cross-sectional population surveys (NMURx, n=8903 and CSEW, n=20 685) with data regarding self-reported recreational drug use and demographics. NMURx is an online survey using non-probability sampling methodology with preset demographical quotas based on census data. CSEW surveys drug use via computer-assisted self-interviewing as part of a computer-assisted personal-interviewing crime survey. RESULTS: Cannabis was the most frequently used drug regardless of demographics. Prevalence of drug use for specific substances was generally higher for males, younger ages and students. The relationship between income and drug misuse is less clear. Self-reported prevalence of drug use in the NMURx survey is consistently higher than CSEW (absolute difference 1%-3 % across substances and timescales) and persists after stratification for gender, age, student status and household income. CONCLUSIONS: The NMURx survey has a broad reach of participants, and a sampling scheme that achieves external validity, compared with general population demographics. NMURx's online format allows flexibility in items surveyed and in response to emerging trends. The self-reported drug use in the NMURx cohort is comparable, although slightly higher, than the CSEW estimates.

Trends in Postpartum Depressive Symptoms - 27 States, 2004, 2008, and 2012.

Postpartum depression is common and associated with adverse infant and maternal outcomes (e.g., lower breastfeeding initiation and duration and poor maternal and infant bonding) (1-3). A developmental Healthy People 2020 objective is to decrease the proportion of women delivering a live birth who experience postpartum depressive symptoms (PDS).* To provide a baseline for this objective, CDC sought to describe self-reported PDS overall, by reporting state, and by selected sociodemographic factors, using 2004, 2008, and 2012 data from the Pregnancy Risk Assessment Monitoring System (PRAMS). A decline in the prevalence of PDS was observed from 2004 (14.8%) to 2012 (9.8%) among 13 states with data for all three periods (p<0.01). Statistically significant (p<0.05) declines in PDS prevalence were observed for eight states, and no significant changes were observed for five states. In 2012, the overall PDS prevalence was 11.5% for 27 states and ranged from 8.0% (Georgia) to 20.1% (Arkansas). By selected characteristics, PDS prevalence was highest among new mothers who 1) were aged ≤19 years or 20-24 years, 2) were of American Indian/Alaska Native or Asian/Pacific Islander race/ethnicity, 3) had ≤12 years of education, 4) were unmarried, 5) were postpartum smokers, 6) had three or more stressful life events in the year before birth, 7) gave birth to term, low-birthweight infants, and 8) had infants requiring neonatal intensive care unit admission at birth. Although the study did not investigate reasons for the decline, better recognition of risk factors for depression and improved screening and treatment before and during pregnancy, including increased use of antidepressants, might have contributed to the decline. However, more efforts are needed to reduce PDS prevalence in certain states and subpopulations of women. Ongoing surveillance and activities to promote appropriate screening, referral, and treatment are needed to reduce PDS among U.S. women.

Risks of Preterm Delivery and Small for Gestational Age Infants: Effects of Nondaily and Low-Intensity Daily Smoking During Pregnancy.

BACKGROUND: Few studies have examined the effects of nondaily smoking or low-intensity daily smoking and infant outcomes. We examined the associations between preterm delivery and small for gestational age (SGA) infants in relation to both nondaily and daily smoking. METHODS: We used population-based data on women who delivered live singleton infants using the 2009-11 Pregnancy Risk Assessment Monitoring System. Women's smoking status in the last 3 months of pregnancy was categorised as nonsmokers, quitters, nondaily smokers (<1 cigarette/day), and daily smokers. Controlling for maternal age, maternal race/ethnicity, education, marital status, prepregnancy body mass index (BMI), trimester of prenatal care entry, parity, and alcohol use, we estimated adjusted prevalence ratios (PR) for the outcomes of preterm delivery (<37 weeks' gestation) and SGA. RESULTS: Of the 88 933 women, 13.1%, 1.7%, and 9.6% of the sample were quitters, nondaily smokers, and daily smokers, respectively, in the last 3 months of pregnancy. While nondaily smoking was not associated with preterm delivery, daily smoking was. However, we found no dose-response relationship with the number of cigarettes smoked per day. Risk of delivering a SGA infant was increased for both nondaily and daily smokers (PR 1.4, 95% CI 1.1, 1.8 and PR 2.0, 95% CI 1.9, 2.2 respectively). CONCLUSIONS: Nondaily smoking in the last 3 months of pregnancy was associated with an increased risk of delivering a SGA infant. Pregnant women should be counselled that smoking, including nondaily and daily smoking, can adversely affect birth outcomes.

Nondaily Smokers' Characteristics and Likelihood of Prenatal Cessation and Postpartum Relapse.

INTRODUCTION: This study aimed to calculate the prevalence of pre-pregnancy nondaily smoking (<1 cigarette/day), risk factors, and report of prenatal provider smoking education; and assess the likelihood of prenatal cessation and postpartum relapse for nondaily smokers. METHODS: We analyzed data from 2009 to 2011 among women with live-born infants participating in the Pregnancy Risk Assessment Monitoring System. We compared characteristics of pre-pregnancy daily smokers (≥1 cigarette/day), nondaily smokers, and nonsmokers (chi-square adjusted p < .025). Between nondaily and daily smokers, we compared proportions of prenatal cessation, postpartum relapse (average 4 months postpartum), and reported provider education. Multivariable logistic regression calculated adjusted prevalence ratios (APR) for prenatal cessation among pre-pregnancy smokers (n = 27 360) and postpartum relapse among quitters (n = 13 577). RESULTS: Nondaily smokers (11% of smokers) were more similar to nonsmokers and differed from daily smokers on characteristics examined (p ≤ .001 for all). Fewer nondaily smokers reported provider education than daily smokers (71.1%, 86.3%; p < .001). A higher proportion of nondaily compared to daily smokers quit during pregnancy (89.7%, 49.0%; p < .001), and a lower proportion relapsed postpartum (22.2%, 48.6%; p < .001). After adjustment, nondaily compared to daily smokers were more likely to quit (APR: 1.65; 95% confidence interval [CI]: 1.58-1.71) and less likely to relapse postpartum (APR: 0.55; 95% CI: 0.48-0.62). CONCLUSIONS: Nondaily smokers were more likely to quit smoking during pregnancy, less likely to relapse postpartum, and less likely to report provider education than daily smokers. Providers should educate all women, regardless of frequency of use, about the harms of tobacco during pregnancy, provide effective cessation interventions, and encourage women to be tobacco free postpartum and beyond. IMPLICATION: Nondaily smoking (<1 cigarette/day) is increasing among US smokers and carries a significant risk of disease. However, smoking patterns surrounding pregnancy among nondaily smokers are unknown. Using 2009-2011 data from the Pregnancy Risk Assessment Monitoring System, we found pre-pregnancy nondaily smokers compared to daily smokers were 65% more likely to quit smoking during pregnancy and almost half as likely to relapse postpartum. Providers should educate all women, regardless of frequency of use, about the harms of tobacco during pregnancy, provide effective cessation interventions, and encourage women to be tobacco free postpartum and beyond.

The Effects of Prepregnancy Body Mass Index and Gestational Weight Gain on Fetal Macrosomia Among American Indian/Alaska Native Women.

OBJECTIVES: The American Indian/Alaska Native (AI/AN) population is a high-risk group across many health indicators, including fetal macrosomia. We aimed to investigate the effects of prepregnancy body mass index (BMI) and gestational weight gain (GWG) on macrosomia and explore possible racial and geographical variations among AI/AN women. METHODS: This retrospective cohort study was conducted from the Pregnancy Risk Assessment Monitoring System in eight states (2004-2011) among live, singleton, term births to AI/AN women 20 years or older. Prevalence of macrosomia (birth weight ≥ 4000 g) by select characteristics were estimated; differences were assessed with Chi-squares. Multivariable logistic regression was conducted to calculate adjusted odds ratios (aOR) for effects on macrosomia of BMI and GWG (enumerating the pounds women deviated from the Institute of Medicine guidelines for GWG) controlling for other factors in the total sample and stratified by race and state of residence. RESULTS: The prevalence of macrosomia was 14 %, ranging from 8 to 21 % (Utah-Alaska). Among AI/AN women, 30 % were obese prepregnancy and 50 % had excess GWG. Significant independent effects were found for macrosomia of prepregnancy overweight (aOR 1.27; 95 % Confidence Interval 1.01-1.59), obesity (aOR 1.63; 1.29-2.07), and excess GWG (aOR 1.16; 1.13-1.20 per five pounds gained beyond appropriate). Adjusted estimates varied between race and state. CONCLUSIONS: Prepregnancy BMI and GWG are independent factors for macrosomia among AI/AN women. Future research should prioritize development, testing, and implementation of weight management programs, which account for variations among AI/AN women, both before and during pregnancy for BMI regulation and GWG control.

The Association Between Diabetes Mellitus Among American Indian/Alaska Native Populations with Preterm Birth in Eight US States from 2004-2011.

OBJECTIVES: Assess risk of preterm birth associated with diabetes mellitus (DM) among American Indian and Alaska Natives (AI/AN), a population with increased risk of DM and preterm birth, and examine whether this association differed by state of residence. METHODS: We used surveillance data from the Pregnancy Risk Assessment Monitoring System from 12,400 AI/AN respondents with singleton births in Alaska, Minnesota, Nebraska, New Mexico, Oklahoma, Oregon, Utah, and Washington from 2004-2011. We conducted multivariable logistic regression models to estimate the odds ratio adjusted for maternal age and prepregnancy BMI with all observations and then stratified by state. RESULTS: DM was reported in 5.92 % of the study population and preterm birth occurred in 8.95 % of births. Women with DM had 1.92 times higher odds of having a preterm birth than women without DM [95 % confidence interval (CI) 1.21-2.78]. After stratifying on state, women with DM in Nebraska had the greatest odds of preterm birth [aOR 6.63, (95 % CI 3.80-11.6)] while women in Alaska saw a protective effect from DM [aOR 0.17, (95 % CI 0.07-0.42)] compared to women without DM. CONCLUSION: Overall, AI/AN women with DM had significantly greater odds of preterm birth compared to AI/AN women without DM across states. Substantial differences in this association between states calls for increased public health efforts in high-risk areas as well as further research to assess whether differences are attributable to diagnosis, reporting, tribal, healthcare or lifestyle factors.

Changes in metabolic acidosis following birth in intensive care unit neonates.

BACKGROUND: Little is known about how and why metabolic acidosis changes within the first six hours of life in intensive care unit neonates. OBJECTIVE: To determine changes in pH and base excess between paired umbilical cord arterial and neonatal arterial blood samples during the first 6 h of life, to identify factors associated with the direction and magnitude of change, and to examine morbidity and mortality in newborns with acidosis at birth or as neonates. STUDY DESIGN: Retrospective cohort study of all deliveries from a single institution between 2016-2020 with paired umbilical cord arterial and neonatal arterial samples obtained within 6 h of life meeting rigorous criteria to ensure sample integrity. The primary outcomes were the direction and magnitude of change of pH and base excess. Multiple factors were assessed for possible correlation with pH and base excess change. The secondary outcome was the association between a composite outcome of death or cerebral palsy and pathologic acidosis (pH ≤ 7.1) at birth or as a neonate. RESULTS: 102 patients met inclusion criteria. Newborn arterial gasses were obtained at a median of 1.5 h (74 % < 2 h). pH improved in 71 % of cases and worsened in 29 %, and base excess improved in 52 % and worsened in 48 %, with wide observed ranges in both parameters. The paired pH and base excess values were moderately (r = 0.38) and strongly (r = 0.63) positively correlated, respectively, but were not correlated with time since birth (r = 0.14). Low birth weight, prematurity or respiratory failure were associated with worsening or less improvement, while worse initial acidosis was associated with greater improvement. Death or survival with cerebral palsy was more common with pathologic acidosis in either cord or newborn sample as compared with those without acidosis (27.3 % vs 3.7 %, p = 0.003), and was more common in those with isolated neonatal acidosis as compared to those without acidosis (50 % vs 3.7 %, p = 0.016). CONCLUSIONS: Changes in pH and base excess occurred over a wide range between delivery and the first newborn blood gas in the first 6 h of life, and we identified several factors associated with direction of change. Metabolic acidosis at birth cannot reliably be inferred from neonatal arterial values. Neonatal acidosis, including acidosis following a normal pH and base excess at birth, was associated with morbidity and mortality.

Clustering patterns in polysubstance mortality in the United States in 2017: a multiple correspondence analysis of death certificate data.

PURPOSE: The main goal of this analysis was to identify mortality patterns apparent when many drug classes are analyzed together. METHODS: The Drug Involved Mortality database is a registry of drug terms mentioned on death certificates of all drug-related deaths in the United States. Means of total number of drugs involved and percentages of specific drug combinations were calculated. Dimensionality reduction using multiple correspondence analysis and hierarchical clustering identified clusters of drugs listed on death certificates. RESULTS: An average of 2.4 specific drugs were listed on death certificates in 2017. For 9 of the top 10 drugs involved, over 80% of deaths involved at least one other drug. As expected, opioid drugs and psychostimulants clustered together, but other psychoactive substances (non-opioid analgesics, sedatives, antidepressants, antipsychotics) clustered together into multi-class groups. Other drugs (e.g., acetaminophen, oxymorphone) were frequently involved in polysubstance death, but did not cluster with any other specific drug. Deaths involving illicit drugs listed fewer drugs than deaths involving prescription drugs. CONCLUSIONS: While individual drug substances might contribute to many deaths (e.g., fentanyl), polysubstance mortality is more common than single substance mortality. Multidimensional analyses integrating all drugs involved are useful to identify uncommon patterns of overdose and changing trends.

The prevalence of blood product transfusion after the implementation of a postpartum hemorrhage bundle: a retrospective cohort at a single safety net academic institution.

BACKGROUND: Postpartum hemorrhage, defined as blood loss of ≥1000 mL within 24 hours after birth, is a leading cause of maternal morbidity and is associated with substantial financial and emotional burden. Based on societal and regulatory guidelines, in 2019, our institution adopted a postpartum hemorrhage prevention and management bundle based on the California Maternal Quality Care Collaborative initiatives. OBJECTIVE: The study aimed to compare the prevalence of maternal blood product transfusion before and after the implementation of the bundle. STUDY DESIGN: This was a retrospective cohort study comparing the prevalence of blood product transfusion before and after the implementation of a California Maternal Quality Care Collaborative-based postpartum hemorrhage management bundle at a single safety net teaching hospital between October 2017 and December 2019 excluding a 4-month rollout period between September 2018 and December 2018. The study included all patients ≥18 years of age and at >20 weeks' gestation. Exclusion criteria were out-born deliveries, delivery at time of significant nonobstetrical trauma, and refusal of blood transfusion. The primary outcome was the frequency of any blood product transfusion in the pre- and postbundle implementation cohorts. Secondary outcomes included blood product transfusion type and amount, maternal death, intrauterine balloon placement, uterine artery embolization, unplanned peripartum hysterectomy, intensive care admission, and length of stay among all deliveries complicated by postpartum hemorrhage. We further evaluated compliance with bundle measures for all postpartum hemorrhage cases. Cohort characteristics were compared using chi-square tests or Fisher exact tests for categorical data and Satterthwaite or Wilcoxon 2-sample tests for continuous variables based on data distributions. The proportion of blood product transfusion were evaluated using a chi-square test. RESULTS: A total of 6744 deliveries were included with 3310 in the pre- and 3425 in the postbundle cohort. The prevalence of any blood product transfusion was similar between the pre- and postbundle cohorts (3.41%; 113/3310 vs 3.47%; 119/3425; P=.892). The prevalence of postpartum hemorrhage was 7.05% (233/3310) in the prebundle cohort and 10.34% (354/3434) in the postbundle cohort (P<.001). Among women with postpartum hemorrhage, those in the prebundle cohort had a higher rate of blood product transfusion than those in the postbundle cohort (36.05%; 84/233 vs 26.84%; 95/354; P=.018). Compared with the prebundle counterparts, patients with postpartum hemorrhage in the postbundle cohort had higher rates of utilization of intrauterine balloon placement (10.30%; 24/233 vs 16.95%; 60/354; P=.024). There were no significant differences among other secondary outcomes. The overall compliance with the bundle among those with blood loss ≥1000 mL was 92.1%. CONCLUSION: The implementation of the postpartum hemorrhage bundle did not decrease the overall prevalence of blood product transfusion and may have led to higher rates of utilization of resources.

Patient Adherence to Follow-Up Recommendations Following Cryotherapy for Treatment of High-Grade Cervical Dysplasia.

Objective Determine the rate of patient adherence to follow-up recommendations after cryotherapy for high-grade cervical lesions, and identify patient characteristics associated with adherence to follow-up. Methods This is a retrospective case series from May 2016 to June 2018 of patients who underwent cryotherapy for high-grade dysplasia at a single academic safety-net hospital. Patient demographics and clinical information were abstracted from the electronic medical record. All patients were recommended to follow up with Pap and high-risk human papillomavirus (HPV) testing 12 months after their procedure. The primary outcome was patient adherence to these recommendations. Descriptive statistics and statistical testing were utilized to compare adherence by demographic and clinical characteristics. A multivariable logistic model was used to preliminarily look at potential factors associated with increased odds of adherence. We further described the proportion of follow-up testing among those patients who adhered to recommendations. Results One hundred and forty-three patients met the inclusion criteria. The adherence percentage was 60.1% (95% CI: 51.6, 68.2). Only employment was associated with follow-up among demographic variables reviewed (p=0.039). Of those who were adherent with follow-up, 4.7% (4/86) had high-grade findings on follow-up Pap testing, and 56.9% (49/86) had negative cytology and negative HPV testing. Conclusion Adherence to follow-up recommendations for the following cryotherapy for high-grade dysplasia within our system was poor, and demographic factors were generally not associated with adherence to follow-up. Given these findings, cryotherapy should be used with caution.

Early pregnancy glycaemia predicts postpartum diabetes mellitus.

OBJECTIVE: To determine the association between early pregnancy glycaemia, as measured by glycosylated haemoglobin A1c (HbA1c) at the first prenatal visit, and persistent postpartum diabetes mellitus (DM). STUDY DESIGN: All women first diagnosed with DM during pregnancy who had both HbA1c prior to 24 weeks and postpartum DM testing were included. The proportions of women with normal (<5.7%), prediabetic (5.7-6.4%) and elevated (≥6.5%) early HbA1c who tested positive for postpartum DM were compared. Test characteristics of HbA1c to predict persistent postpartum DM were calculated. RESULTS: One hundred and twenty-one women met the study inclusion criteria. HbA1c was obtained at a median gestational age of 9 weeks. Twenty-two women (18.2%) had persistent postpartum DM, which was highly correlated with early HbA1c: 16 (73%) women had an elevated HbA1c, five (22.7%) women had a prediabetic HbA1c and only one (4.5%) woman had a normal HbA1c. Of 65 women with gestational DM and a normal early HbA1c, only one (1.5%) had persistent DM within the first year (negative predictive value 98.5%). Sixteen of 18 women with an elevated early HbA1c had persistent postpartum DM (positive predictive value 88.9%). These percentages were significant overall and between groups (p < 0.001). No clinical or demographic factors were highly predictive of postpartum DM. CONCLUSIONS: Early pregnancy glycaemia, as measured by HbA1c at the first prenatal visit, is highly predictive of persistent postpartum DM, and may allow clinically important risk stratification to prioritize postpartum testing and care. Postpartum DM is rare amongst women with gestational DM who begin the pregnancy with a normal HbA1c, while postpartum DM is highly likely for those with an elevated HbA1c in early pregnancy. Nearly three-quarters of women who tested positive for DM post partum had an elevated HbA1c in early pregnancy, indicating that they had undiagnosed DM prior to conception.

Development and validation of a prediction model for postpartum hemorrhage at a single safety net tertiary care center.

BACKGROUND: Postpartum hemorrhage is a leading cause of pregnancy-related morbidity and mortality; however, there is limited ability to identify women at risk of this obstetrical complication. OBJECTIVE: This study aimed to develop and validate a prediction model for postpartum hemorrhage based on antenatal and intrapartum risk factors. STUDY DESIGN: This was a retrospective cohort study of women who delivered between April 2016 and March 2019 at a single safety net hospital. The prevalence of postpartum hemorrhage, defined as blood loss of ≥1000 mL at the time of delivery, was determined, and characteristics were compared between women with and without postpartum hemorrhage. Women were randomly assigned to a prediction or a validation cohort. The selection of predictors to be included in the model was based on known antenatal and intrapartum risk factors for postpartum hemorrhage. A multivariable logistic regression with a backward stepwise approach was used to create a prediction model. Area under the receiver operating characteristic curve and 95% bootstrap confidence intervals were calculated. Using the final model, a single threshold for classifying postpartum hemorrhage was chosen, and the resulting sensitivity, specificity, and false-negative and false-positive rates were explored. RESULTS: The prevalence rates of postpartum hemorrhage in the prediction and validation cohorts were 6.3% (377 of 6000 cases) and 6.4% (241 of 3774 cases), respectively (P=.83). The following predictors were selected for the final model: maternal body mass index (kg/m2), number of fetuses, history of postpartum hemorrhage, admission platelets of <100,000/µL, chorioamnionitis, arrest of descent, placental abruption, and active labor duration. The predictive model had an area under the receiver operating characteristic curve of 0.82 (95% confidence interval, 0.81-0.84). When applied to the validation cohort, the model had an area under the receiver operating characteristic curve of 0.81 (95% confidence interval, 0.78-0.83), a sensitivity of 86.9%, a specificity of 74.2%, a positive predictive value of 18.6%, a negative predictive value of 98.8%, a false-negative rate of 13.1%, and a false-positive rate of 25.9%. CONCLUSION: The model performed reasonably well in identifying women at risk of postpartum hemorrhage. Further studies are necessary to evaluate the model in clinical practice and its effect on decreasing the prevalence of postpartum hemorrhage and associated maternal morbidity.

Prevalence and description of kratom (Mitragyna speciosa) use in the United States: a cross-sectional study.

BACKGROUND AND AIMS: Mitragyna speciosa ('kratom') contains mu opioid partial agonists. It is widely available, and occasionally used as a home remedy for opioid use disorder. The Drug Enforcement Agency considers kratom a drug of concern; however, prevalence of use and role in drug misuse are unknown. This study aimed to characterize kratom use in the United States. DESIGN: Cross-sectional Survey of Non-Medical Use of Prescription Drugs (NMURx) Program, 2018 third quarter and 2019 first quarter. SETTING: A validated non-probability online survey in the United States. PARTICIPANTS: A total of 59 714 respondents aged 18 years or older, weighted to represent the adult US population (n = 252 063 800). MEASUREMENTS: In addition to prevalence of past-year kratom and other drug use, behavior proportions were estimated. The Drug Abuse Screening Test (DAST-10) estimated consequences of drug abuse. FINDINGS: The estimated prevalence of past-year kratom use in the adult US population was 0.8% [95% confidence interval (CI) = 0.7-0.9], representing 2 031 803 adults. Life-time prevalence was 1.3% (95% CI = 1.2-1.4), representing 3 353 624 adults. Kratom users were younger (mean 35 years, P < 0.001), with higher proportions of males (61.0 versus 48.6%, P < 0.001), students (14.1 versus 7.5%, P < 0.001) and health-care professionals (9.7 versus 4.5%, P < 0.001) and fewer bachelor's/advanced degree graduates (33.4 versus 42.6%, P < 0.001) compared with non-users. Results were inconclusive on whether there was a difference in kratom use by race, household income or employment status. Among those with past-year kratom use, 36.7% (95% CI = 32.1-41.3) non-medically used prescription opioids, 21.7% (95% CI = 18.0-25.5) used illicit opioids, 54.4% (95% CI = 49.5-59.3) used another illicit drug and 67.1% (95% CI = 62.5-71.8) used cannabis. The DAST-10 profile was more often substantial/severe in kratom users (21 versus 1%, P < 0.001) compared with non-users. CONCLUSIONS: Estimated United States past-year prevalence of kratom use is 0.8%, and kratom users tend to have more serious substance abuse profiles than non-users or users of cannabis, alcohol or cigarettes. To our knowledge, this is the first description of kratom use at the national level.