Comparison of six statistical methods for interrupted time series studies: empirical evaluation of 190 published series

Epidemiological characteristics and interrupted time series analysis of mumps in Quzhou City, 2005-2023

To analyze epidemiological characteristics of mumps in different periods and explore the impact of different immunization strategies of mumps containing vaccine (MuCV) on the incidence of mumps in Quzhou. We collected reported cases of mumps in Quzhou from 2005 to 2023 for descriptive epidemiological analysis. Interrupted time series (ITS) analysis was used to assess changes in mumps incidence during the one-dose, two-dose, and the three-dose MuCV vaccination period. From 2005 to 2023, a total of 10,295 cases of mumps were reported. The average annual reported incidence rates during three period, were 57.5/100,000 (7015 cases), 14.8/100,000 (2590 cases), and 6/100,000 (690 cases), respectively. The male-to-female ratio reported cases was 1.58:1, and most cases were aged 5-9 years, accounting for 43.7%. ITS analysis showed an immediate decrease in monthly mumps incidence of 3.3951/100,000 after two-dose MuCV vaccination period (p = .022). However, the monthly mumps incidence only decreased by 0.1191/100,000 immediately after the implementation of the three doses of MMR vaccination (p = .411). The monthly incidence of mumps showed a slight increase of 0.0052/100,000 in the one-dose vaccination period, followed by a decrease of 0.0126/100,000 in the two-dose vaccination period, and a further decrease of 0.0004/100,000 in the three-dose MMR period (p > .05). This study shows that timely introduction of two doses of MuCV and ensuring high vaccination coverage are beneficial in controlling mumps epidemics. Currently, the key populations affected are children and adolescents, emphasizing the importance of addressing outbreaks within school epidemics.

The impact of COVID-19 restrictions on crime and its implications for public health: a case study from a small developing country

BACKGROUND

Crime has been described as a public health issue in Trinidad and Tobago, a small developing nation. COVID-19 restrictions, aimed at maintaining public health safety by limiting disease spread, may have negatively impacted crime rates due to an alteration of social and economic conditions. This study evaluates the implications of these restrictions on crime dynamics and hence their impacts on overall public health.

METHODS

Employing interrupted time series analysis with seasonal autoregressive integrated moving average with exogenous factor (Seasonal Autoregressive Integrated Moving Average with Exogenous Variables) models, monthly data on murders, sexual offences, and motor vehicle larcenies from January 2013 to June 2023 were analysed.

RESULTS

The study found a decrease in murders and motor vehicle larcenies with the onset of restrictions, followed by an inverse trend correlating with the easing of measures. Sexual offences showed no significant change in response to the restrictions.

CONCLUSIONS

While COVID-19 restrictions initially influenced certain crime rates, the effect varied across crime types. Policy interventions based on lessons learned from the COVID-19 pandemic must be done to strategically reduce and prevent crime without having the negative side effects of the pandemic.

Road to Recovery: protocol for a mixed-methods prospective cohort study evaluating the impact of a new model of substance use care in a Canadian setting

INTRODUCTION

The Road to Recovery (R2R) Initiative is an innovative model of substance use care that seeks to increase treatment capacity by creating approximately 100 new addiction treatment beds to provide on-demand addiction care in Vancouver, British Columbia, for patients with substance use disorders. The new model also coordinates the region's existing clinical substance use services to support patients across a care continuum that includes traditional office-based addiction treatment and harm reduction services, early withdrawal management and more intensive abstinence-based treatment programming. To understand the impact of offering on-demand and coordinated substance use care, an observational cohort of individuals who access any R2R clinical service will be created to examine health and social outcomes over time.

METHODS AND ANALYSIS

This prospective mixed-methods study will invite individuals from Vancouver, Canada, who access substance use treatment through the R2R model of care to (1) complete a baseline and 12-month follow-up quantitative questionnaire that solicits sociodemographic, substance use and previous addiction treatment data and (2) provide consent to the use of participants' personal identifiers to access health records for chart review and for annual linkage to select health and administrative databases to allow for ongoing (virtual) community follow-up over 5 years. Additionally, a purposive sample of cohort participants will be invited to participate in baseline and 12-month follow-up qualitative interviews to share their experiences accessing R2R and identify challenges and opportunities associated with the implementation of R2R.

ETHICS AND DISSEMINATION

The study was approved by the University of British Columbia Providence Health Care Research Ethics Board in September 2023. Results from the proposed study will be published in peer-reviewed journals, presented at national and international scientific conferences and disseminated through regular meetings with policymakers, individuals with lived and living experience, and other high-level stakeholders, academic presentations and lay media.

The impact of meropenem shortage and post-prescription review and feedback on broad-spectrum antimicrobial use: An interrupted time-series analysis

Background

Meropenem (MEPM) holds significance in treating severe infections and drug-resistant bacteria. There are concerns that antimicrobial shortages may lead to the use of alternative antimicrobials that are less effective and safer. We have responded to the MEPM shortage with post-prescription monitoring and feedback (PPRF) with no restrictions on MEPM initiation. We aimed to assess the impact of the MEPM shortage and the PPRF on broad-spectrum antimicrobial use and mortality.

Methods

This retrospective study was conducted in a single hospital in Japan. The period from October 2021 to August 2022 was defined as the period before the MEPM shortage, and the period from September 2022 to March 2023 was defined as the period during the MEPM shortage. To support the appropriate use of antimicrobials during MEPM shortages, the antimicrobial stewardship team (AST) developed a list of alternatives to MEPM. An interrupted time series analysis was used to assess changes in use and mortality among patients receiving broad-spectrum antimicrobials over the study period.

Discussion

The shortage of MEPM and PPRF temporarily increased the use of alternative cefepime; however, the subsequent change in days of therapy and days of coverage of broad-spectrum antimicrobials suggests a decrease in the use of these antimicrobials. Despite these shifts, the mortality rates remained stable, suggesting that the response to the shortage did not adversely affect treatment outcomes.

Conclusion

In the context of antimicrobial shortages, AST support plays an important role in enabling physicians to make optimal use of antimicrobials.

Differences in Cesarean Rates for Nulliparous, Term, Singleton, Vertex Births Among Racial and Ethnic Groups and States Before and After Stay-at-Home Orders During the COVID-19 Pandemic, United States, 2017-2021

OBJECTIVES

Evidence is limited on differences in cesarean rates for nulliparous, term, singleton, vertex (NTSV) births across racial and ethnic groups at the national and state level during the COVID-19 pandemic. We assessed changes in levels and trends of NTSV cesarean rates before and after stay-at-home orders (SAHOs) were implemented in the United States (1) overall, (2) by racial and ethnic groups, and (3) by 50 US states from January 2017 through December 2021.

METHODS

We used birth certificate data from 2017 through 2021, restricted to hospital births, to calculate monthly NTSV cesarean rates for the United States and for racial and ethnic groups and to calculate quarterly NTSV cesarean rates for the 50 states. We used interrupted time-series analysis to measure changes in NTSV cesarean rates before and after implementation of SAHOs (March 1 through May 31, 2020).

RESULTS

Of 6 022 552 NTSV hospital births, 1 579 645 (26.2%) were cesarean births. Before implementation of SAHOs, NTSV cesarean rates were declining in the United States overall; were declining among births to non-Hispanic Asian, non-Hispanic Black, Hispanic, and non-Hispanic White women; and were declining in 6 states. During the first month of implementation of SAHOs in May 2020, monthly NTSV rates increased in the United States by 0.55%. Monthly NTSV rates increased by 1.20% among non-Hispanic Black women, 0.90% among Hispanic women, and 0.28% among non-Hispanic White women; quarterly NTSV rates increased in 6 states.

CONCLUSION

In addition to emergency preparedness planning, hospital monitoring, and reporting of NTSV cesarean rates to increase provider awareness, reallocation and prioritization of resources may help to identify potential strains on health care systems during public health emergencies such as the COVID-19 pandemic.

Strategies for Implementing GlobalConsent to Prevent Sexual Violence in University Men (SCALE): study protocol for a national implementation trial

BACKGROUND

Globally, women 15-24 years are at heightened risk of sexual violence victimization, a risk factor for adverse mental, physical, and behavioral health outcomes. Sexual violence is common at universities and most often perpetrated by men, yet few evidence-based prevention strategies targeting men have been tested in low- and middle-income countries. GlobalConsent is a six-module, web-based educational program adapted from an efficacious U.S.-based program. Nine months post-treatment in a randomized trial in Vietnam, GlobalConsent reduced men's sexually violent behavior (odds ratio [OR] = 0.71, 95%CI 0.50-1.00) and increased prosocial intervening behavior (OR = 1.51, 1.00-2.28) relative to an attention-control. Evidence regarding optimal implementation strategies for scale up is needed.

METHODS

We will randomize six medical universities in North, Central, and South Vietnam to deliver GlobalConsent using two different packages of implementation strategies that vary in intensity. Higher-intensity strategies will include greater (1) pre- and post-implementation engagement with university leaders and faculty and (2) greater pre-implementation outreach, follow-up, and incentives for students to promote engagement and completion of GlobalConsent. Higher intensity universities will receive additional training and support for their added activities. We will compare implementation drivers and outcomes, intervention effectiveness, and cost-effectiveness across the two implementation bundles. Our mixed-methods comparative interrupted time series design includes (1) qualitative interviews and quantitative surveys with university leaders and implementation teams to assess implementation barriers and facilitators; (2) repeated surveys with leaders and faculty, implementation teams, and male students to assess multilevel implementation drivers and outcomes; (3) repeated surveys with male students to assess behavioral outcomes (sexual violence and intervening behavior) and mediating variables (knowledge, attitudes, affect, and capacities); and (4) time diaries and cost tracking to assess cost-effectiveness of the two implementation-strategies bundles.

DISCUSSION

This project is the first to assess packages of implementation strategies to deliver an efficacious web-based sexual violence prevention program for undergraduate men across all regions of Vietnam and synergizes with a violence-prevention training initiative (D43TW012188). This approach will produce rigorous evidence about how to disseminate GlobalConsent nationally, which holds promise to reduce gender-based health inequities linked to sexual violence as GlobalConsent is brought to scale.

TRIAL REGISTRATION

NCT06443541. Retrospectively registered with ClinicalTrials.gov. Registered on June 05, 2024.

The impact of the English calorie labelling policy on the energy content of food offered and purchased in worksite cafeterias: a natural experiment

BACKGROUND

On 6 April 2022, legislation came into effect in England requiring calorie labels to be applied to food items on menus of larger food businesses. This study aimed to assess the impact of calorie labelling on (a) food purchased and (b) energy content of menu options in worksite cafeterias.

METHODS

Product-level sales data and energy content of available items was obtained from 142 worksite cafeterias from January 2022-October 2022. Interrupted-time-series (ITS) analysis with level and slope change evaluated daily energy (kcal) purchased per item, and ITS with level change assessed mean energy per option available on menus before and after calorie labelling. Each analysis was conducted 6 weeks and 6 months from implementation. A post-hoc ITS examined weekly energy purchased per item over a longer period (March 2021-October 2022; 135 sites).

RESULTS

There was no evidence calorie labelling changed the energy content of foods purchased (6-week: + 0.60 cal/product, 95%CI:-2.54, + 3.75; 6-month: + 1.59 cal/product, 95%CI:-0.96, + 4.16). Post-hoc analyses suggested calorie labels were associated with a reduction in mean energy of items purchased over time (-0.65 kcal/week, 95%CI:-0.81,-0.49), but a significant increase (+ 3 kcal, 95%CI: + 0.43, + 5.60) at the point of implementation. There was a reduction in the mean energy content of menu options at each seasonal menu change (April 2022:-1.79 kcal, 95%CI:-3.42,-0.15; July 2022:-4.18 kcal, 95% CI:-7.65,-0.73).

CONCLUSION

This large observational study in worksite cafeterias found no evidence to indicate the introduction of calorie labelling led to any immediate reduction in energy purchased by customers. There was some evidence of increasing impact over time, possibly associated with changes in menu offerings, but this effect was small and cannot be directly attributed to calorie labelling.

Women's health in focus: Real-world data on valproate prescriptions during pregnancy - a cohort study in Catalonia (Spain)

OBJECTIVES

To characterise the exposure to valproate within a cohort of pregnant women using electronic health records (EHRs) from Catalonia (System for the Development of Research in Primary Care, SIDIAP).

DESIGN

Drug-utilisation cohort study covering the period from January 2011 to June 2020. The study included pregnancy episodes of women from Catalonia identified by the algorithm.

SETTING

Data were sourced from SIDIAP, a comprehensive EHR repository that includes information from various data sources: recorded prescriptions (both hospital and primary care), diagnoses and sociodemographic characteristics identified by primary care physicians, and sexual and reproductive health data from ASSIR (used by gynaecologists and midwives).

PARTICIPANTS

Women aged 12-50 with at least one pregnancy episode occurred during January 2011-June 2020 and at least a prescription of valproate during pregnancy.

PRIMARY AND SECONDARY OUTCOMES

Primary outcomes included valproate exposure, measured through prevalence and cumulative incidence in pregnancy episodes and by trimester. The impact of regulatory measures (risk mitigation measures, RMMs) was assessed, and prescriptions over time were analysed using interrupted time series analysis. Secondary outcomes included health issues, pregnancy outcomes, smoking habits and socioeconomic characteristics.

RESULTS

A total of 99 605 pregnancies were identified, with at least 3.03‰ (95% CI 2.69‰ to 3.39‰) exposed to valproate at some point (302 pregnancies, 276 women). The median pregnancy duration was 38.30 weeks (IQR 12.6-40.1), and the median age at pregnancy was 32.37 years (IQR 27.20-36.56). Epilepsy was the most frequent health issue. The prevalence and cumulative incidence of valproate prescriptions decreased during pregnancy and increased postpregnancy. The RMMs implemented in 2014 led to a reduction in monthly valproate prescriptions during pregnancy in this cohort.

CONCLUSIONS

The study highlights the decline in valproate prescriptions during pregnancy due to RMMs and underscores the need for standardised methodologies in future studies to ensure the safety of pregnant patients and optimise scientific evidence.

The performance of interrupted time series designs with a limited number of time points: Learning losses due to school closures during the COVID-19 pandemic

Interrupted time series (ITS) designs are increasingly used for estimating the effect of shocks in natural experiments. Currently, ITS designs are often used in scenarios with many time points and simple data structures. This research investigates the performance of ITS designs when the number of time points is limited and with complex data structures. Using a Monte Carlo simulation study, we empirically derive the performance-in terms of power, bias and precision- of the ITS design. Scenarios are considered with multiple interventions, a low number of time points and different effect sizes based on a motivating example of the learning loss due to COVID school closures. The results of the simulation study show the power of the step change depends mostly on the sample size, while the power of the slope change depends on the number of time points. In the basic scenario, with both a step and a slope change and an effect size of 30% of the pre-intervention slope, the required sample size for detecting a step change is 1,100 with a minimum of twelve time points. For detecting a slope change the required sample size decreases to 500 with eight time points. To decide if there is enough power researchers should inspect their data, hypothesize about effect sizes and consider an appropriate model before applying an ITS design to their research. This paper contributes to the field of methodology in two ways. Firstly, the motivation example showcases the difficulty of employing ITS designs in cases which do not adhere to a single intervention. Secondly, models are proposed for more difficult ITS designs and their performance is tested.

Self-Harm and Suicide Rates Before and After an Early Intervention Program for Patients With First-Episode Schizophrenia

Importance

Evidence on the association of early intervention services (EISs) with self-harm and suicide among patients with first-episode schizophrenia (FES) at older than 25 years is lacking.

Objective

To examine changes in self-harm and suicide rates among patients with FES before and after the implementation of an EIS program.

Design, Setting, and Participants

This population-based cohort study conducted among 37 040 patients aged 15 to 64 years with FES between January 1, 2001, and March 31, 2020, used electronic medical records from the Hong Kong Clinical Data Analysis and Reporting System. All patients were followed up from the first diagnosis of schizophrenia (the index date) until the date of their death or the end of the study period (March 31, 2021), whichever came first. Statistical analysis was performed from July to November 2023.

Exposure

The EIS extended the Early Assessment Service for Young People With Early Psychosis (EASY) program from patients aged 15 to 25 years to those aged 15 to 64 years (EASY Plus). The exposure was the implementation of the EASY Plus program in April 2011. The exposure period was defined as between April 2012 and March 2021 for the 1-year-time-lag analysis.

Main Outcomes and Measures

The outcomes were monthly rates of self-harm and suicide among patients with FES before and after the implementation of the EASY Plus program. Interrupted time series analysis was used for the main analysis.

Results

This study included 37 040 patients with FES (mean [SD] age at onset, 39 [12] years; 82.6% older than 25 years; 53.0% female patients). The 1-year-time-lag analysis found an immediate decrease in self-harm rates among patients aged 26 to 44 years (rate ratio [RR], 0.77 [95% CI, 0.59-1.00]) and 45 to 64 years (RR, 0.70 [95% CI, 0.49-1.00]) and among male patients (RR, 0.71 [95% CI, 0.56-0.91]). A significant long-term decrease in self-harm rates was found for all patients with FES (patients aged 15-25 years: RR, 0.98 [95% CI, 0.97-1.00]; patients aged 26-44 years: RR, 0.98 [95% CI, 0.97-0.99]; patients aged 45-64 years: RR, 0.97 [95% CI, 0.96-0.98]). Suicide rates decreased immediately after the implementation of the EASY Plus program among patients aged 15 to 25 years (RR, 0.33 [95% CI, 0.14-0.77]) and 26 to 44 years (RR, 0.38 [95% CI, 0.20-0.73]). Compared with the counterfactual scenario, the EASY Plus program might have led to 6302 fewer self-harm episodes among patients aged 26 to 44 years.

Conclusions and Relevance

This cohort study of the EASY Plus program suggests that the extended EIS was associated with reduced self-harm and suicide rates among all patients with FES, including those older than 25 years. These findings emphasize the importance of developing tailored interventions for patients across all age ranges to maximize the benefits of EISs.

Epidemiology of brief resolved unexplained events and impact of clinical practice guidelines in general and pediatric emergency departments

OBJECTIVES

The aim of this study was to describe the incidence of brief resolved unexplained events (BRUEs) and compare the impact of a national clinical practice guideline (CPG) on admission and diagnostic testing practices between general and pediatric emergency departments (EDs).

METHODS

Using the Nationwide Emergency Department Sample for 2012-2019, we conducted a cross-sectional study of children <1 year of age with an International Classification of Diseases diagnostic code for BRUE. Population incidence rate was estimated using Centers for Disease Control and Prevention birth data. ED incidence rate was estimated for all ED encounters. We used interrupted time series to evaluate the associated impact of the CPG publication on the outcomes of ED disposition (discharge, admission, and transfer) and electrocardiogram (ECG) use.

RESULTS

Of 133,972 encounters for BRUE, 80.0% occurred in general EDs. BRUE population incidence was 4.28 per 1000 live births and the annual incidence remained stable (p = 0.19). BRUE ED incidence was 5.06 per 1000 infant ED encounters (p = 0.14). The impact of the BRUE CPG on admission rates was limited to pediatric EDs (level shift -23.3%, p = 0.002). Transfers from general EDs did not change with the CPG (level shift 2.2%, p = 0.17). After the CPG was published, ECGs increased by 13.7% in pediatric EDs (p = 0.005) but did not change in general EDs (level shift -0.2%, p = 0.82).

CONCLUSIONS

BRUEs remain a common pediatric problem at a population level and in EDs. Although a disproportionate number of infants present to general EDs, there is differential uptake of the CPG recommendations between pediatric and general EDs. These findings may support quality improvement opportunities aimed at improving care for these infants and decreasing unnecessary hospital admissions or transfers.

Impact of the COVID-19 pandemic on the Italian national viral hepatitis surveillance: an interrupted time series analysis, 2006-2022

OBJECTIVES

This study aimed to assess the impact of the COVID-19 pandemic on national surveillance of viral hepatitis in Italy.

STUDY DESIGN

Interrupted time series analysis.

METHODS

Using an interrupted time series analysis with a customised AutoRegressive Integrated Moving Average model on hepatitis cases reported to the Integrated Epidemiological System of Acute Viral Hepatitis from 2006 to 2022, we examined trends in incidence, time to diagnosis and time to epidemiological investigations for hepatitis A, hepatitis B and hepatitis C.

RESULTS

The quarterly incidence of hepatitis B (-0.251, P = 0.05) and hepatitis C (-0.243, P = 0.003) significantly decreased with the onset of the pandemic. Surveillance times for hepatitis B (5.496, P < 0.001) and hepatitis C (35.723, P < 0.001), measured as days lost per quarter, significantly increased 12 and 24 months after the pandemic's start. This aligns with a notable rise in quarterly incidence at 36 months for both (0.032, P = 0.030 for hepatitis B; 0.040, P < 0.001 for hepatitis C).

CONCLUSIONS

The decrease in reported cases could be due to an increase in infection prevention control and containment measures put in place in a pandemic context. However, a delay in the initiation of epidemiological investigations was observed, which could lead to a further increase in incidence in the future.

Impact of an improved outdoor space on people with dementia in a hospital unit

Introduction

Gardens and outdoor spaces are an important part of institutional environments for people with dementia. However, evidence regarding the benefits these spaces have for people with dementia is still limited. This paper presents the evaluation of the redevelopment of an inaccessible outdoor space into a therapeutic garden on a high dependency psychogeriatric unit in an acute hospital.

Method

A Mixed methods evaluation was undertaken. An interrupted time series analysis investigated the impact of the garden on falls and challenging behaviours of patients using routinely collected data. Perspectives of the redeveloped garden were captured through (a) a staff survey and (b) semi-structured interviews with families of patients.

Results

Rates of falls and challenging behaviours dropped at the time of the garden opening but showed increasing rates each month both before and after the garden opened. Most staff believed that the garden provided benefits for patients however limited staff time and concerns over patient safety were barriers to use. Families identified four main themes related to garden use including: (1) being outside (2) occupation and identity, (3) being stimulating, and (4) barriers and facilitators.

Conclusion

The garden was regarded positively by families and staff however, there were barriers that prevented it from being better utilised. Staff concerns over risk were not reflected in falls and challenging behaviour outcomes. Further research into how barriers to garden use may be overcome is justified.

The Impact of an Antimicrobial Stewardship Program on Days of Therapy in the Pediatric Center: An Interrupted Time-Series Analysis of a 19-Year Study

BACKGROUND

We aimed to analyze the effects of an antimicrobial stewardship program (ASP) on the proportion of antimicrobial-resistant pathogens in bacteremia, antimicrobial use, and mortality in pediatric patients.

METHODS

A retrospective single-center study was performed on pediatric inpatients under 19 years old who received systemic antimicrobial treatment from 2001 to 2019. A pediatric infectious disease attending physician started ASP in January 2008. The study period was divided into the pre-intervention (2001-2008) and the post-intervention (2009-2019) periods. The amount of antimicrobial use was defined as days of therapy per 1,000 patient-days, and the differences were compared using delta slope (= changes in slopes) between the two study periods by an interrupted time-series analysis. The proportion of resistant pathogens and the 30-day overall mortality rate were analyzed by the χ².

RESULTS

The proportion of extended-spectrum beta-lactamase (ESBL)-producing Escherichia coli and Klebsiella pneumoniae bacteremia increased from 17% (39 of 235) in the pre-intervention period to 35% (189 of 533) in the post-intervention period (P < 0.001). The total amount of antimicrobial use significantly decreased after the introduction of ASP (delta slope value = -16.5; 95% confidence interval [CI], -30.6 to -2.3; P = 0.049). The 30-day overall mortality rate in patients with bacteremia did not increase, being 10% (55 of 564) in the pre-intervention and 10% (94 of 941) in the post-intervention period (P = 0.881).

CONCLUSION

The introduction of ASP for pediatric patients reduced the delta slope of the total antimicrobial use without increasing the mortality rate despite an increased incidence of ESBL-producing gram-negative bacteremia.

Impact of capitation on physicians' behavior among patients with hypertension: an interrupted time series study in rural China

OBJECTIVE

The purpose of this study is to explore the change in physicians' hypertension treatment behavior before and after the reform of the capitation in county medical community.

METHODS

Spanning from January 2014 to December 2019, monthly data of outpatient and inpatient were gathered before and after the implementation of the reform in April 2015. We employed interrupted time series analysis method to scrutinize the instantaneous level and slope changes in the indicators associated with physicians' behavior.

RESULTS

Several indicators related to physicians' behavior demonstrated enhancement. After the reform, medical cost per visit for inpatient exhibited a reverse trajectory (-53.545, 95%CI: -78.620 to -28.470, p < 0.01). The rate of change in outpatient drug combination decelerated (0.320, 95%CI: 0.149 to 0.491, p < 0.01). The ratio of infusion declined for both outpatient and inpatient cases (-0.107, 95%CI: -0.209 to -0.004, p < 0.1; -0.843, 95%CI: -1.154 to -0.532, p < 0.01). However, the results revealed that overall medical cost per visit and drug proportion for outpatient care continued their initial upward trend. After the reform, the decline of drug proportion for outpatient care was less pronounced compared to the period prior to the reform, and length of stay also had a similar trend.

CONCLUSION

To some extent, capitation under the county medical community encourages physicians to control the cost and adopt a more standardized diagnosis and treatment behavior. This study provides evidence to consider the impact of policy changes on physicians' behavior when designing payment methods and healthcare systems aimed at promoting PHC.

Consumption of psychotropic drugs in Croatia before and during the COVID-19 pandemic: a 10-year longitudinal study (2012-2021)

PURPOSE

This longitudinal study aimed to examine the trends in antipsychotics, antidepressants, anxiolytics, and hypnotics/sedatives consumption in Croatia over a 10-year period (2012-2021). The study also assessed whether the COVID-19 pandemic had an impact on the yearly consumption of psychotropic drugs.

METHODS

Data were collected from Croatian Agency for Medicinal Products and Medical Devices (HALMED) and presented as defined daily doses per 1000 inhabitants per day (DDD/TID). The consumption before (2012-2019) and during the COVID-19 pandemic (2020-2021) was compared with interrupted time series analysis.

RESULTS

There was an increase in total consumption of analyzed psychotropic drugs in Croatia between the years 2012 and 2021, from 115.47 DDD/TID in 2012 to 155.50 DDD/TID in 2021. An increasing trend was observed in the consumption of all 4 analyzed groups of medicines (antipsychotics, anxiolytics, hypnotics and sedatives, and antidepressants). Anxiolytics accounted for 59% (68.29/115.47 DDD/TID), and hypnotics and sedatives for 8.5% (9.76/115.47 DDD/TID) of total consumption in 2012. At the end of a 10-year period, hypnotics and sedatives represented 12% (19.05/155.50 DDD/TID) and anxiolytics 54% (83.53/155.50 DDD/TID) of psychotropic drugs consumption. The total consumption of psychotropic drugs was not significantly different before and during COVID-19 pandemic (estimate ± standard error = 5.029 ± 6.899, t = 0.729, P = 0.490).

CONCLUSION

Croatia had a high, continuously increasing consumption of psychotropic drugs. National anxiolytics consumption was one of the highest globally, while consumption of antidepressants was rather low compared to other high-income countries. The COVID-19 pandemic did not seem to influence the yearly utilization of psychotropic drugs in Croatia.

Impact of the COVID-19 pandemic on delivery of prostate cancer care in Australia: An interrupted time series analysis

The COVID-19 pandemic led to a major disruption to health services across the world. The aim of this population-based study was to assess the downstream effects of the pandemic on diagnostic tests and treatment activities related to prostate cancer (PC). The Australian Government Department of Health Medicare Benefits Schedule and the Pharmaceutical Benefits Scheme databases were queried from January 2010 to June 2022. Two interrupted time series were performed Pre-COVID (January 2010 to February 2020) and peri-COVID (March 2020 to June 2022). Temporal modeling was performed to account for seasonal variation. Pre-COVID-19, monthly prostate-specific antigen (PSA) testing showed a declining trend and testing decreased by 81 tests per 100 000 annually. A single-month 38% drop in PSA testing was observed in April 2020; this corresponded to Australia's first wave. No change was observed in the rate of prostate biopsies. Peri-COVID-19 outbreaks, there was a slight shift toward the use of long-acting androgen deprivation therapy (ADT) at 4% with a predilection still for short-acting agents. with no registered change in the overall volume of radiotherapy or surgery. There were no deficits in the number of diagnostic and treatment activities for men with PC. Aside from a slight shift toward long-acting ADT use during the pandemic, no other patterns were observed. The longer-term impact such as missed diagnosis or late presentation affecting chances of survival due to COVID-19 is yet to be ascertained.

Design and statistical analysis reporting among interrupted time series studies in drug utilization research: a cross-sectional survey

INTRODUCTION

Interrupted time series (ITS) design is a commonly used method for evaluating large-scale interventions in clinical practice or public health. However, improperly using this method can lead to biased results.

OBJECTIVE

To investigate design and statistical analysis characteristics of drug utilization studies using ITS design, and give recommendations for improvements.

METHODS

A literature search was conducted based on PubMed from January 2021 to December 2021. We included original articles that used ITS design to investigate drug utilization without restriction on study population or outcome types. A structured, pilot-tested questionnaire was developed to extract information regarding study characteristics and details about design and statistical analysis.

RESULTS

We included 153 eligible studies. Among those, 28.1% (43/153) clearly explained the rationale for using the ITS design and 13.7% (21/153) clarified the rationale of using the specified ITS model structure. One hundred and forty-nine studies used aggregated data to do ITS analysis, and 20.8% (31/149) clarified the rationale for the number of time points. The consideration of autocorrelation, non-stationary and seasonality was often lacking among those studies, and only 14 studies mentioned all of three methodological issues. Missing data was mentioned in 31 studies. Only 39.22% (60/153) reported the regression models, while 15 studies gave the incorrect interpretation of level change due to time parameterization. Time-varying participant characteristics were considered in 24 studies. In 97 studies containing hierarchical data, 23 studies clarified the heterogeneity among clusters and used statistical methods to address this issue.

CONCLUSION

The quality of design and statistical analyses in ITS studies for drug utilization remains unsatisfactory. Three emerging methodological issues warranted particular attention, including incorrect interpretation of level change due to time parameterization, time-varying participant characteristics and hierarchical data analysis. We offered specific recommendations about the design, analysis and reporting of the ITS study.

A Comprehensive Analysis of the Impact of the COVID-19 Pandemic on Lung Cancer in Delaware

Background

COVID-19 has greatly impacted the U.S. health system. What is not as well-understood is how this has altered specific aspects of lung cancer care. While cancer incidence and screening have been affected, it is not known whether pre-existing racial and socioeconomic disparities worsened or if treatment standards changed. The purpose of this study is to provide a comprehensive analysis of the impact of COVID-19 on lung cancer in the state of Delaware.

Methods

Health care claims were analyzed from the Delaware Healthcare Claims Database for the years 2019-2020. Patients with a new lung cancer diagnosis and those who had undergone lung cancer screening were identified. Demographic and socioeconomic variables including gender, age, race, and insurance were studied. Patients were analyzed for type of treatment by CPT code. The intervention of interest in this study was the institution of restrictions at the end of March 2020. An interrupted time series analysis (ITSA) was utilized to evaluate baseline levels and overall trend changes.

Results

The incidence of lung cancer diagnoses and lung cancer screenings decreased in the nine-month time period after the initiation of COVID-19 lockdowns. Demographic and socioeconomic variables including gender, race, income, and education level were not affected; however, statistical differences were seen in the most elderly subgroup. Treatment modalities including number of surgeries, chemotherapy, and radiation therapy did not change significantly.

Conclusions

COVID-19 has had a significant impact on lung cancer care within the state of Delaware. Lung cancer incidence, screenings, and elderly patients were affected the most.

Interpretation of coefficients in segmented regression for interrupted time series analyses

Background Segmented regression, a common model for interrupted time series (ITS) analysis, primarily utilizes two equation parametrizations. Interpretations of coefficients vary between the two segmented regression parametrizations, leading to occasional user misinterpretations. Methods To illustrate differences in coefficient interpretation between two common parametrizations of segmented regression in ITS analysis, we derived analytical results and present an illustration evaluating the impact of a smoking regulation policy in Italy using a publicly accessible dataset. Estimated coefficients and their standard errors were obtained using two commonly used parametrizations for segmented regression with continuous outcomes. We clarified coefficient interpretations and intervention effect calculations. Results Our investigation revealed that both parametrizations represent the same model. However, due to differences in parametrization, the immediate effect of the intervention is estimated differently under the two approaches. The key difference lies in the interpretation of the coefficient related to the binary indicator for intervention implementation, impacting the calculation of the immediate effect. Conclusions Two common parametrizations of segmented regression represent the same model but have different interpretations of a key coefficient. Researchers employing either parametrization should exercise caution when interpreting coefficients and calculating intervention effects.

Comparison of statistical methods used to meta-analyse results from interrupted time series studies: an empirical study

BACKGROUND

The Interrupted Time Series (ITS) is a robust design for evaluating public health and policy interventions or exposures when randomisation may be infeasible. Several statistical methods are available for the analysis and meta-analysis of ITS studies. We sought to empirically compare available methods when applied to real-world ITS data.

METHODS

We sourced ITS data from published meta-analyses to create an online data repository. Each dataset was re-analysed using two ITS estimation methods. The level- and slope-change effect estimates (and standard errors) were calculated and combined using fixed-effect and four random-effects meta-analysis methods. We examined differences in meta-analytic level- and slope-change estimates, their 95% confidence intervals, p-values, and estimates of heterogeneity across the statistical methods.

RESULTS

Of 40 eligible meta-analyses, data from 17 meta-analyses including 282 ITS studies were obtained (predominantly investigating the effects of public health interruptions (88%)) and analysed. We found that on average, the meta-analytic effect estimates, their standard errors and between-study variances were not sensitive to meta-analysis method choice, irrespective of the ITS analysis method. However, across ITS analysis methods, for any given meta-analysis, there could be small to moderate differences in meta-analytic effect estimates, and important differences in the meta-analytic standard errors. Furthermore, the confidence interval widths and p-values for the meta-analytic effect estimates varied depending on the choice of confidence interval method and ITS analysis method.

CONCLUSIONS

Our empirical study showed that meta-analysis effect estimates, their standard errors, confidence interval widths and p-values can be affected by statistical method choice. These differences may importantly impact interpretations and conclusions of a meta-analysis and suggest that the statistical methods are not interchangeable in practice.

The Association Between the License Fee Increase and the Density of Tobacco Retailers in California-A Segmented Interrupted Time-Series Analysis by Income and Race/Ethnicity

INTRODUCTION

On May 9, 2016, the State of California passed a law to increase the licensing fee for tobacco retailers from a one-time-only fee of $100 to an annual fee of $265, effective on June 9, 2016. This study investigates the association between this fee increase and retailer densities by neighborhood income and race/ethnicity characteristics.

METHODS

We obtained quarterly data on the number of active tobacco retailer licenses from 2011 to 2020 in every zip code in California from the California Department of Tax and Fee Administration. These data were then linked to zip code-level income, race/ethnicity, and population measures. We used a single-group segmented interrupted time-series analysis to assess the association between the increase in licensing fees and retailer densities by neighborhood income and race/ethnicity.

RESULTS

After the implementation of the annual licensing fees, the retailer density decreased both immediately and gradually. Specifically, the retailer density dropped by 0.47 in the first quarter following the intervention. Compared to the pre-intervention time trend, the retailer density decreased quarterly by 0.05. Furthermore, the impacts of increasing licensing fees were more pronounced in low-income and the majority Black zip codes.

CONCLUSIONS

Given that higher smoking prevalence is associated with greater tobacco outlet density, the licensing fee increase could be an effective policy tool to reduce tobacco use among economically disadvantaged and minority Black communities, thereby addressing tobacco-use disparities.

IMPLICATIONS

This study used the single-group segmented interrupted time-series analysis to assess the association between the licensing fee increase and tobacco retailer densities by neighborhood income and race/ethnicity. We found that this licensing fee increase was associated with reduced retailer densities and the total number of active retailers right after the implementation. We further found that the annual licensing fee policy had a continuous effect in reducing tobacco retailer densities in all zip codes. The impacts of increasing licensing fees were more pronounced in low-income and majority of Black zip codes.

Impact of COVID-19 on patterns of drug utilization: A case study at national hospital

BACKGROUND

The Coronavirus disease of 2019 (COVID-19) pandemic and the corresponding mitigation measures have had a discernible impact on drug utilization among outpatients. However, limited research exists on the prescription trends in the elderly population during the pandemic period in Viet Nam.

OBJECTIVES

This study aims to analyze the effects of COVID-19 on outpatient drug utilization patterns at a national geriatric hospital in Ho Chi Minh City before and after the early onset of the pandemic.

METHODS

Data was collected from the prescriptions and administration claims, encompassing the period from January 2016 to December 2022. The dataset was divided into two periods: Period 1: January 2016 to December 2020 and Period 2: January 2021 to December 2022. The drug utilization was measured using DDD/1000P (defined daily doses-DDD per 1000 prescriptions) on a monthly basis. The analysis employed interrupted time series using Autoregressive Integrated Moving Average (ARIMA) to detect changes in drug use levels and rates.

RESULTS

A total of 1,060,507 and 644,944 outpatient prescriptions from Thong Nhat Hospital were included in Period 1 and Period 2, respectively. The median age of the patients were 58 in Period 1 and 67 years old in Period 2. The most common comorbidities were dyslipidemia, hypertension, and diabetes mellitus. In terms of medication utilization, cardiovascular drugs were the most frequently prescribed, followed by drugs active on the digestive and hormonal systems. The study observed significant surges in the number of prescriptions and the average number of drugs per prescription. However, there were no significant changes in the overall consumption of all drugs. Among the drug groups related to the cardiovascular system, three subgroups experienced a sudden and significant increase: cardiac therapy, beta-blocking agents, and antihypertensives, with increasing consumption levels of 1,177.73 [CI 95%: 79.29; 2,276.16], 73.32 [CI 95%: 28.18; 118.46], and 36.70 [CI 95%: 6.74; 66.66] DDD/1000P, respectively. On the other hand, there was a significant monthly decrease of -31.36 [CI 95%: -57.02; -5.70] DDD/1000P in the consumption of anti-inflammatory and antirheumatic products. Interestingly, there was a significant increase of 74.62 [CI 95%: -0.36; 149.60] DDD/1000P in the use of antigout preparations.

CONCLUSION

COVID-19 resulted in a sudden, non-significant increase in overall drug consumption levels among outpatients. Notably, our findings highlight significant increases in the utilization of three drug groups related to the cardiovascular system, specifically cardiac therapy, beta-blocking agents, and antihypertensives. Intriguingly, there was a statistically significant increase in the consumption of antigout preparations, despite a decline in the monthly consumption rate of non-steroidal anti-flammatory drugs (NSAIDs). Further studies in the following years are necessary to provide a more comprehensive understanding of the impact of COVID-19 on outpatient drug utilization patterns.

The impact of mass vaccination policy and control measures on lumpy skin disease cases in Thailand: insights from a Bayesian structural time series analysis

Introduction

In 2021, Thailand reported the highest incidence of lumpy skin disease (LSD) outbreaks in Asia. In response to the widespread outbreaks in cattle herds, the government's livestock authorities initiated comprehensive intervention measures, encompassing control strategies and a national vaccination program. Yet, the efficacy of these interventions remained unevaluated. This research sought to assess the nationwide intervention's impact on the incidence of new LSD cases through causal impact analysis.

Methods

Data on weekly new LSD cases in Thailand from March to September 2021 was analyzed. The Bayesian structural time series (BSTS) analysis was employed to evaluate the causal relationship between new LSD cases in the pre-intervention phase (prior to the vaccination campaign) and the post-intervention phase (following the vaccination campaign). The assessment involved two distinct scenarios, each determined by the estimated effective intervention dates. In both scenarios, a consistent decline in new LSD cases was observed after the mass vaccination initiative, while other control measures such as the restriction of animal movement, insect control, and the enhancement of the active surveillance approach remained operational throughout the pre-intervention and the post-intervention phases.

Results and discussion

According to the relative effect results obtained from scenario A and B, it was observed that the incidence of LSD cases exhibited reductions of 119% (95% Credible interval [CrI]: -121%, -38%) and 78% (95% CrI: -126, -41%), respectively. The BSTS results underscored the significant influence of these interventions, with a Bayesian one-sided tail-area probability of p < 0.05. This model-based study provides insight into the application of BSTS in evaluating the impact of nationwide LSD vaccination based on the national-level data. The present study is groundbreaking in two respects: it is the first study to quantify the causal effects of a mass vaccination intervention on the LSD outbreak in Thailand, and it stands as the only endeavor of its kind in the Asian context. The insights collected from this study hold potential value for policymakers in Thailand and other countries at risk of LSD outbreaks.

On the role of benchmarking data sets and simulations in method comparison studies

Method comparisons are essential to provide recommendations and guidance for applied researchers, who often have to choose from a plethora of available approaches. While many comparisons exist in the literature, these are often not neutral but favor a novel method. Apart from the choice of design and a proper reporting of the findings, there are different approaches concerning the underlying data for such method comparison studies. Most manuscripts on statistical methodology rely on simulation studies and provide a single real-world data set as an example to motivate and illustrate the methodology investigated. In the context of supervised learning, in contrast, methods are often evaluated using so-called benchmarking data sets, that is, real-world data that serve as gold standard in the community. Simulation studies, on the other hand, are much less common in this context. The aim of this paper is to investigate differences and similarities between these approaches, to discuss their advantages and disadvantages, and ultimately to develop new approaches to the evaluation of methods picking the best of both worlds. To this aim, we borrow ideas from different contexts such as mixed methods research and Clinical Scenario Evaluation.

COVID-19 pandemic and initiation of treatment for atrial fibrillation: a nationwide analysis of claims data

BACKGROUND

The COVID-19 pandemic profoundly disrupted the delivery of medical care. It remains unclear whether individuals diagnosed with new onset disease during the pandemic were less likely to initiate treatments after diagnosis. We sought to evaluate changes in the treatment initiation of patients newly diagnosed with atrial fibrillation (AF) after the onset of the COVID-19 pandemic.

METHODS

In this retrospective cohort study, we identified individuals with incident AF from 01/01/2016-09/30/2021 using Optum's de-identified Clinformatics® Data Mart Database. The primary outcome was initiation of oral anticoagulation (OAC) within 30 days of AF diagnosis. Secondary outcomes included initiation of OAC within 180 days of diagnosis, initiation of warfarin, direct oral anticoagulants (DOACs), rhythm control medications and electrical cardioversion within 30 days of diagnosis. We constructed interrupted time series analyses to examine changes in the outcomes following the onset of the pandemic.

RESULTS

A total of 573,524 patients (age 73.0 ± 10.9 years) were included in the study. There were no significant changes in the initiation of OAC, DOAC, and rhythm control medications associated with the onset of the pandemic. There was a significant decrease in initiation of electrical cardioversion associated with the onset of the pandemic. The rate of electronic cardioversion within 30 days of diagnosis decreased by 4.9% per 1,000 patients after the onset of the pandemic and decreased by about 35% in April 2020, compared to April 2019, from 5.53% to 3.58%.

CONCLUSION

The COVID-19 pandemic did not affect the OAC initiation within 30 days of AF diagnosis but was associated with a decline in the provision of procedures for patients newly diagnosed with AF.

Impact of Risk Minimisation Measures on Valproate Use among Women of Reproductive Age in Latvia Between 2013 and 2020: A 7-Year Nationwide Prescription Database Study

BACKGROUND

A relevant safety concern for the use of valproate (VPA) in women of reproductive age is its teratogenicity. In 2014 European Medicines Agency (EMA) introduced risk minimisation measures (RMMs) to reduce the VPA use by women of reproductive age, where the impact on VPA use was not as large as expected. In 2018, the EMA introduced additional RMMs, and it is essential to assess impact of these interventions.

OBJECTIVE

The objective of this study was to evaluate the impact of the EMA-published RMMs in 2014 and 2018 on the prevalence of VPA use and to describe trends in the prevalence rate and incidence proportion of VPA use in epilepsy, bipolar disorder and off-label indications in Latvia.

METHODS

This was a nationwide population-based study using a primary care prescription database. The study included women in age groups < 15, 15-49 and > 49 years and men in age group 15-49 years who have received VPA. This study assessed the prevalence rate and the incidence proportion of VPA use. The impact of RMMs on the two study intervention periods [fourth quarter (Q4) 2014 and Q4 2018] in men and women was evaluated using causal impact analysis.

RESULTS

In the study cohort, VPA use in women in the age group 15-49 years decreased after the first and second intervention periods, where after the first intervention period the relative reduction in prevalence of VPA consumption was -7.7 [95% confidence interval (CI) -10%, -5.1%] and after both study periods -6.4% (95% CI -11%, -1.5%). In girls < 15 years of age, valproate use decreased after both intervention periods, while in women > 49 years old VPA use increased. In men aged 15-49 years, an increase after the first period and a non-significant decrease after both intervention periods was observed. The prevalence of valproate use in girls < 15 years and women 15-49 years of age with bipolar disorder, epilepsy and off-label indications decreased per 1000 people during the study period. The incidence proportion of VPA use in women aged 15-49 years decreased each year since the beginning of the study period.

CONCLUSIONS

A statistically significant decrease in the prevalence of VPA use was identified among girls < 15 years and women 15-49 years of age. In Latvia, an overall good reaction to the EMA RMMs was observed. The effects go beyond the target population and affect the use of VPA in young girls as well.

Resilience-Informed Community Violence Prevention and Community Organizing Strategies for Implementation: Protocol for a Hybrid Type 1 Implementation-Effectiveness Trial

BACKGROUND

Community violence is a persistent and challenging public health problem. Community violence not only physically affects individuals, but also its effects reverberate to the well-being of families and entire communities. Being exposed to and experiencing violence are adverse community experiences that affect the well-being and health trajectories of both children and adults. In the United States, community violence has historically been addressed through a lens of law enforcement and policing; the impact of this approach on communities has been detrimental and often ignores the strengths and experiences of community members. As such, community-centered approaches to address violence are needed, yet the process to design, implement, and evaluate these approaches is complex. Alternatives to policing responses are increasingly being implemented. However, evidence and implementation guidance for community-level public health approaches remain limited. This study protocol seeks to address community violence through a resilience framework-Adverse Community Experiences and Resilience (ACE|R)-being implemented in a major US city and leveraging a strategy of community organizing to advance community violence prevention.

OBJECTIVE

The objective of this research is to understand the impact of community-level violence prevention interventions. Furthermore, we aim to describe the strategies of implementation and identify barriers to and facilitators of the approach.

METHODS

This study uses a hybrid type 1 effectiveness-implementation design. Part 1 of the study will assess the effectiveness of the ACE|R framework plus community organizing by measuring impacts on violence- and health-related outcomes. To do so, we plan to collect quantitative data on homicides, fatal and nonfatal shootings, hospital visits due to nonaccidental injuries, calls for service, and other violence-related data. In Part 2 of the study, to assess the implementation of ACE|R plus community organizing, we will collect process data on community engagement events, deliver community trainings on community leadership and organizing, and conduct focus groups with key partners about violence and violence prevention programs in Milwaukee.

RESULTS

This project received funding on September 1, 2020. Prospective study data collection began in the fall of 2021 and will continue through the end of 2023. Data analysis is currently underway, and the first results are expected to be submitted for publication in 2024.

CONCLUSIONS

Community violence is a public health problem in need of community-centered solutions. Interventions that center community and leverage community organizing show promise in decreasing violence and increasing the well-being of community members. Methods to identify the impact of community-level interventions continue to evolve. Analysis of outcomes beyond violence-specific outcomes, including norms and community beliefs, may help better inform the short-term and proximal impacts of these community-driven approaches. Furthermore, hybrid implementation-effectiveness trials allow for the inevitable contextualization required to disseminate community interventions where communities drive the adaptations and decision-making.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/50444.

Impact of the Choose Well Contraceptive Access Initiative on Method Use Among Women Enrolled in South Carolina's Medicaid Program: A Mid-line Assessment

INTRODUCTION

A six-year statewide contraceptive access initiative focused on equitable access to contraception, removing cost barriers, capacity building and training, raising consumer awareness, and expanding contraceptive care at safety net clinics was implemented in South Carolina beginning in 2017. This study assessed changes in contraceptive method use among women enrolled in the South Carolina Medicaid program during the first three years of Choose Well.

METHODS

Contraception use among a retrospective cohort of women aged 15 to 45 enrolled in South Carolina Medicaid from 2012 to 2020 was examined. Interrupted time series regression analysis was used to assess changes in the use of intrauterine devices (IUDs) and contraceptive implants between 2012 and 2016 and 2017 and 2020. Analyses were conducted for all women and stratified by age groups.

RESULTS

Long-acting reversible contraception use increased from 8.5% during the pre-Choose Well period to 10.9% during the Choose Well period (p < .001), with IUD use increasing from 4.3% to 5.2% (p < .001) and implant use increasing from 4.6% to 6.0% (p < .001). The interrupted time series analysis found a significant positive change in the average level of monthly IUD use after Choose Well began (0.493 percentage points; 95% confidence interval, 0.311-0.675). The effect was stronger among women 20 to 25 years of age. Choose Well significantly increased the trend in IUD use among all women by a positive 0.013 percentage points (95% confidence interval, 0.006-0.020) per month beyond expected values.

CONCLUSIONS

At the mid-point of the Choose Well Evaluation, the use of IUD methods increased significantly beyond what would be expected had pre-Choose Well trends continued. This was particularly evident among women 20 to 25 years of age. These findings suggest that Choose Well succeeded in reducing barriers to the use of IUDs.

Changes in Cannabis-Attributable Hospitalizations Following Nonmedical Cannabis Legalization in Canada

Importance

The impact of adult-use cannabis legalization and subsequent commercialization (ie, increasing store and product access) on hospitalizations in Canada is unclear.

Objectives

To examine changes in overall and subtypes of hospitalizations due to cannabis and associated factors following legalization in Canada and to compare changes between provinces.

Design, Setting, and Participants

This repeated cross-sectional analysis included all acute hospitalizations for individuals aged 15 to 105 years in Canada's 4 most populous provinces (Ontario, Quebec, Alberta and British Columbia, population 26.9 million individuals in 2018). Data were obtained from routinely collected health administrative databases. Immediate and gradual changes in the age- and sex-standardized rates of hospitalizations due to cannabis were compared using an interrupted time series design over 3 time periods: prelegalization (January 2015 to September 2018), legalization with product and store restrictions (October 2018 to February 2020), and commercialization, which overlapped with the COVID-19 pandemic (March 2020 to March 2021).

Main Outcomes and Measures

Rates of hospitalizations due to cannabis per 100 000 individuals and per 1000 all-cause hospital admissions.

Results

There were 105 203 hospitalizations due to cannabis over the 7-year study period, 69 192 of which (65.8%) were among male individuals, and 34 678 (33%) of which were among individuals aged 15 to 24 years. Overall, the age- and sex-standardized rate of hospitalizations increased 1.62 times between January 2015 (3.99 per 100 000 individuals) and March 2021 (6.46 per 100 000 individuals). The largest relative increase in hospitalizations was for cannabis-induced psychosis (rate ratio, 1.40; 95% CI, 1.34 to 1.47 during the commercialization period relative to the prelegalization period). Nationally, legalization with restrictions was associated with a gradual monthly decrease of -0.06 (95% CI -0.08 to -0.03) in hospitalizations due to cannabis per 100 000 individuals. Commercialization and the COVID-19 pandemic were associated with an immediate increase of 0.83 (95% CI, 0.30 to 1.30) hospitalizations due to cannabis per 100 000 individuals. There was provincial variation in changes, with provinces with less mature legal markets experiencing the greatest declines immediately following legalization.

Conclusions and Relevance

This cross-sectional study found that legalization with restrictions was not associated with an increase in hospitalizations due to cannabis but commercialization was. The findings suggest that commercialization of cannabis may be associated with increases in cannabis-related health harms, including cannabis-induced psychosis.

Regional differences in the utilisation of antenatal care and skilled birth attendant services during the COVID-19 pandemic in Nigeria: an interrupted time series analysis

INTRODUCTION

The COVID-19 pandemic has had a substantial negative impact on the utilisation of essential health services (EHS) globally, especially in resource-limited settings such as Nigeria. High maternal deaths associated with low access to and utilisation of EHS such as antenatal care (ANC) and skilled birth attendants (SBAs) remain a concern during the COVID-19 era. The study assessed the COVID-19 pandemic effects on ANC and SBA utilisation across regions in Nigeria.

METHODS

Monthly data on ANC and SBA between January 2017 and July 2021 were obtained from the Federal Ministry of Health database. An interrupted time-series analysis, implemented using the Prophet model, was conducted to compare the regional variation of outcomes during the COVID-19 pandemic. Average percentage changes (PC) between the observed and predicted outcomes including their 95% CI were reported.

RESULTS

From March 2020 to July 2021, the number of ANC visits was significantly lower than expected by a 16%-43% change in five of the six regions in Nigeria. The highest significant reduction was in North-West (PC=-43.4; 95% CI: -52.6 to -34.1) and the least in South-West (PC=-15.5; 95% CI: -24.8 to -6.1), with no significant change in the South-East. The number of deliveries by SBA was significantly lower than expected by a 18%-43% change in all the regions (p<0.01). North-East (PC=-43.3; 95% CI: -51.7 to -34.9) and South-West (PC=-18.3; 95% CI: -25.2 to -11.5), respectively, had the highest and the least decline in SBA utilisation. Overall, ANC and SBA patterns of change were relatively similar across the north-south divide though the change effect was considerably pronounced in the north.

CONCLUSION

There was a substantial reduction in ANC and SBA utilisation due to the COVID-19 pandemic in Nigeria, especially in the northern regions. Targeted and contextually relevant interventions should be implemented to alleviate the impact of emergency response on access to EHS and promote access to care during the pandemic.

Comparative Analysis of Three Predictive Models of Performance Indicators with Results-Based Management: Cancer Data Statistics in a National Institute of Health

Predictive models play a crucial role in RBMs to analyze performance indicator results to manage unexpected events and make timely decisions to resolve them. Their use in Mexico is deficient, and monitoring and evaluation are among the weakest pillars of the model. In response to these needs, the aim of this study was to perform a comparative analysis of three predictive models to analyze 10 medical performance indicators and cancer data related to children with cancer. To accomplish these purposes, a comparative and retrospective study with nonprobabilistic convenience sampling was conducted. The predictive models were exponential smoothing, autoregressive integrated moving average, and linear regression. The lowest mean absolute error was used to identify the best model. Linear regression performed best regarding nine of the ten indicators, with seven showing p < 0.05. Three of their assumptions were checked using the Shapiro-Wilk, Cook's distance, and Breusch-Pagan tests. Predictive models with RBM are a valid and relevant instrument for monitoring and evaluating performance indicator results to support forecasting and decision-making based on evidence and must be promoted for use with cancer data statistics. The place numbers obtained by cancer disease inside the main causes of death, morbidity and hospital outpatients in a National Institute of Health were presented as evidence of the importance of implementing performance indicators associated with children with cancer.

The impact of universal health coverage and COVID-19 pandemic on out-of-pocket expenses in Thailand: an analysis of household survey from 1994 to 2021

OBJECTIVES

After Thailand achieved Universal Health Coverage (UHC) in 2002, the extent of financial risk protection has not been assessed in the long term, especially after the COVID-19 pandemic. Therefore, this study aims to revisit the impact of UHC on out-of-pocket expenses (OOPE) for health and to descriptively explore the impact of COVID-19 on OOPE.

METHODS

This study was a secondary data analysis and used data from the Socio-Economic Survey from 1994 to 2021 in Thailand. The effect of UHC on the percentage of OOPE in total health expenditures (THE) from 1994 to 2019 was investigated with an interrupted time-series analysis. Descriptive analyses of OOPE in absolute value during the COVID-19 were conducted.

RESULTS

The percentage of OOPE in THE significantly decreased both before (β -2.02%; 95% CI: -2.70% to - 1.33%) and during (β 1.41%; 95% CI: 0.70% to 2.11%) the UHC period. During the pandemic, total household OOPE for medical equipment was found to have rapidly increased from 643 million THB in 2019 to 9.4 billion THB in 2020.

CONCLUSIONS

The trend of providing financial risk protection (measured by OOPE/THE) in Thailand continues until 2019. Providing medical equipment in sufficient and equally accessible manners should be prioritized during the future pandemic.

Effect estimates can be accurately calculated with data digitally extracted from interrupted time series graphs

Interrupted time series (ITS) studies are frequently used to examine the impact of population-level interventions or exposures. Systematic reviews with meta-analyses including ITS designs may inform public health and policy decision-making. Re-analysis of ITS may be required for inclusion in meta-analysis. While publications of ITS rarely provide raw data for re-analysis, graphs are often included, from which time series data can be digitally extracted. However, the accuracy of effect estimates calculated from data digitally extracted from ITS graphs is currently unknown. Forty-three ITS with available datasets and time series graphs were included. Time series data from each graph was extracted by four researchers using digital data extraction software. Data extraction errors were analysed. Segmented linear regression models were fitted to the extracted and provided datasets, from which estimates of immediate level and slope change (and associated statistics) were calculated and compared across the datasets. Although there were some data extraction errors of time points, primarily due to complications in the original graphs, they did not translate into important differences in estimates of interruption effects (and associated statistics). Using digital data extraction to obtain data from ITS graphs should be considered in reviews including ITS. Including these studies in meta-analyses, even with slight inaccuracy, is likely to outweigh the loss of information from non-inclusion.

Incidence and trends in workplace violence within emergency departments in the United Kingdom 2017-2022: an observational time series analysis

Background

Workplace violence (WPV) is a notable issue facing healthcare services and workers globally. WPV impacts upon the well-being of staff and can put healthcare provision at risk with detrimental effects on patient care. This study aims to investigate and quantify, at national and regional levels, the incidence and trends of WPV within emergency departments (EDs).

Methods

We requested data relating to WPV from all 152 trusts with an ED in the United Kingdom from January 2017-March 2022. We applied interrupted time series and trend analysis to check for significant differences in WPV across the COVID-19 pandemic.

Results

We conducted time series analysis on 58 million attendances and detected statistically significant increases in WPV in March 2020-5.06/100,000 attendances (95% CI 1.59/100,000-8.53/100,000 p < 0.01) and May 2020-20.63/100,000 attendances (95% CI 9.39-31.87 p < 0.01). Rises in incidents of 0.37/100,000 attendances per month (95% CI 0.21-0.53 p < 0.0001) were found January 2017-March 2020. We analyzed 96 million attendances for yearly trends, which revealed statistically significant increasing trends of WPV in London and North-West England (p < 0.05), and physical WPV in the North West England (p < 0.05).

Conclusion

There have been dramatic increases in incidents of WPV in United Kingdom EDs over the last 5 years with concerning rises during the COVID-19 period. Our findings highlight the potential to further demoralize a workforce already under significant strain, resulting in increased absences for physical or mental health and an exodus of staff. Therefore, trusts should ensure there are robust systems in place to protect and safeguard staff.

Impact assessment of a raw coal ban on maternal and child health outcomes in Ulaanbaatar: a protocol for an interrupted time series study

INTRODUCTION

Despite a decade of policy actions, Ulaanbaatar's residents continue to be exposed to extreme levels of air pollution, a major public health concern, especially for vulnerable populations such as pregnant women and children. In May 2019, the Mongolian government implemented a raw coal ban (RCB), prohibiting distribution and use of raw coal in households and small businesses in Ulaanbaatar. Here, we present the protocol for an interrupted time series (ITS; a strong quasi-experimental study design for public health interventions) that aims to assess the effectiveness of this coal ban policy on environmental (air quality) and health (maternal and child) outcomes.

METHODS AND ANALYSIS

Routinely collected data on pregnancy and child respiratory health outcomes between 2016 and 2022 in Ulaanbaatar will be collected retrospectively from the four main hospitals providing maternal and/or paediatric care as well as the National Statistics Office. Hospital admissions data for childhood diarrhoea, an unrelated outcome to air pollution exposure, will be collected to control for unknown or unmeasured coinciding events. Retrospective air pollution data will be collected from the district weather stations and the US Embassy. An ITS analysis will be conducted to determine the RCB intervention impact on these outcomes. Prior to the ITS, we have proposed an impact model based on a framework of five key factors, which were identified through literature search and qualitative research to potentially influence the intervention impact assessment.

ETHICS AND DISSEMINATION

Ethical approval has been obtained via the Ministry of Health, Mongolia (No.445) and University of Birmingham (ERN_21-1403). To inform relevant stakeholders of our findings, key results will be disseminated on both (inter)national and population levels through publications, scientific conferences and community briefings. These findings are aimed to provide evidence for decision-making in coal pollution mitigation strategies in Mongolia and similar settings throughout the world.

In praise of Prais-Winsten: An evaluation of methods used to account for autocorrelation in interrupted time series

Interrupted time series are increasingly being used to assess the population impact of public health interventions. These data are usually correlated over time (auto correlated) and this must be accounted for in the analysis. Typically, this is done using either the Prais-Winsten method, the Newey-West method, or autoregressive-moving-average (ARMA) modeling. In this paper, we illustrate these methods via a study of pneumococcal vaccine introduction and explore their performance under 20 simulated autocorrelation scenarios with sample sizes ranging between 20 and 300. We show that in terms of mean square error, the Prais-Winsten and ARMA methods perform best, while in terms of coverage the Prais-Winsten method generally performs better than other methods. All three methods are unbiased. As well as having good statistical properties, the Prais-Winsten method is attractive because it is decision-free and produces a single measure of autocorrelation that can be compared between studies and used to guide sample size calculations. We would therefore encourage analysts to consider using this simple method to analyze interrupted time series.

The Impact of Opioid Prescribing Limits on Drug Usage in South Carolina: A Novel Geospatial and Time Series Data Analysis

The opioid crisis in the United States has had devastating effects on communities across the country, leading many states to pass legislation that limits the prescription of opioid medications in an effort to reduce the number of overdose deaths. This study investigates the impact of South Carolina's prescription limit law (S.C. Code Ann. 44-53-360), which aims to reduce opioid overdose deaths, on opioid prescription rates. The study utilizes South Carolina Reporting and Identification Prescription Tracking System (SCRIPTS) data and proposes a distance classification system to group records based on proximity and evaluates prescription volumes in each distance class. Prescription volumes were found to be highest in classes with pharmacies located further away from the patient. An Interrupted Time Series (ITS) model is utilized to assess the policy impact, with benzodiazepine prescriptions as a control group. The ITS models indicate an overall decrease in prescription volume, but with varying impacts across the different distance classes. While the policy effectively reduced opioid prescription volumes overall, an unintended consequence was observed as prescription volume increased in areas where prescribers were located at far distances from patients, highlighting the limitations of state-level policies on doctors. These findings contribute to the understanding of the effects of prescription limit laws on opioid prescription rates and the importance of considering location and distance in policy design and implementation.

Primary school mobile dental program in New South Wales, Australia: protocol for the evaluation of a state government oral health initiative

BACKGROUND

Socioeconomically disadvantaged children are disproportionately affected by oral disease. Mobile dental services help underserved communities overcome barriers to accessing health care, including time, geography, and trust. The NSW Health Primary School Mobile Dental Program (PSMDP) is designed to provide diagnostic and preventive dental services to children at their schools. The PSMDP is mainly targeted toward high-risk children and priority populations. This study aims to evaluate the program's performance across five local health districts (LHDs) where the program is being implemented.

METHODS

The evaluation will use routinely collected administrative data, along with other program-specific data sources, from the district public oral health services to conduct a statistical analysis that determines the reach and uptake of the program, its effectiveness, and the associated costs and cost-consequences. The PSMDP evaluation program utilises data from Electronic Dental Records (EDRs) and other data sources, including patient demographics, service mix, general health, oral health clinical data and risk factor information. The overall design includes cross-sectional and longitudinal components. The design combines comprehensive output monitoring across the five participating LHDs and investigates the associations between socio-demographic factors, service patterns and health outcomes. Time series analysis using difference-in-difference estimation will be conducted across the four years of the program, involving services, risk factors, and health outcomes. Comparison groups will be identified via propensity matching across the five participating LHDs. An economic analysis will estimate the costs and cost-consequences for children who participate in the program versus the comparison group.

DISCUSSION

The use of EDRs for oral health services evaluation research is a relatively new approach, and the evaluation works within the limitations and strengths of utilising administrative datasets. The study will also provide avenues to improve the quality of data collected and system-level improvements to better enable future services to be aligned with disease prevalence and population needs.

The impact of COVID-19 on the number of active small primary care businesses by severity of the pandemic: evidence from South Korea

BACKGROUND

Health systems have become financially fragile owing to the economic recession caused by the COVID-19 pandemic. However, small primary care businesses have received less policy attention than public health and secondary care. We aimed to estimate the impact of COVID-19 on the number of active small primary care businesses in South Korea.

METHODS

We selected clinics, dental clinics, oriental clinics, and pharmacies as primary care businesses. Our estimation took advantage of regional variations in COVID-19 cases in South Korea. We determined the number of active primary care businesses from 2019 1Q to 2021 1Q on a quarterly basis, and conducted interrupted time series analysis to estimate the effects of COVID-19 on this sector.

RESULTS

This study found no significant increase or decrease in the number of clinics, dental clinics, and oriental clinics immediately after the pandemic began or in the time trends after the pandemic. However, there was a significant increase in the number of pharmacies immediately after the pandemic. The most affected area presented different trends in the number of pharmacies, dental clinics, and oriental clinics.

CONCLUSIONS

Impact of the pandemic on the number of active small primary care business were low in South Korea. However, the impact varied according to the type of primary care setting and severity of the pandemic. The additional public health role of primary care could be associated with the sustenance of primary care businesses.

Evaluation of statistical methods used in the analysis of interrupted time series studies: a simulation study

BACKGROUND

Interrupted time series (ITS) studies are frequently used to evaluate the effects of population-level interventions or exposures. However, examination of the performance of statistical methods for this design has received relatively little attention.

METHODS

We simulated continuous data to compare the performance of a set of statistical methods under a range of scenarios which included different level and slope changes, varying lengths of series and magnitudes of lag-1 autocorrelation. We also examined the performance of the Durbin-Watson (DW) test for detecting autocorrelation.

RESULTS

All methods yielded unbiased estimates of the level and slope changes over all scenarios. The magnitude of autocorrelation was underestimated by all methods, however, restricted maximum likelihood (REML) yielded the least biased estimates. Underestimation of autocorrelation led to standard errors that were too small and coverage less than the nominal 95%. All methods performed better with longer time series, except for ordinary least squares (OLS) in the presence of autocorrelation and Newey-West for high values of autocorrelation. The DW test for the presence of autocorrelation performed poorly except for long series and large autocorrelation.

CONCLUSIONS

From the methods evaluated, OLS was the preferred method in series with fewer than 12 points, while in longer series, REML was preferred. The DW test should not be relied upon to detect autocorrelation, except when the series is long. Care is needed when interpreting results from all methods, given confidence intervals will generally be too narrow. Further research is required to develop better performing methods for ITS, especially for short series.