Daily temperature variability and mental health-related hospital visits in New York State

BACKGROUND

Despite plausible behavioral and physiological pathways, limited evidence exists on how daily temperature variability is associated with acute mental health-related episodes.

OBJECTIVES

We aimed to explore associations between daily temperature range (DTR) and mental health-related hospital visits using data of all hospital records in New York State during 1995-2014. We further examined factors that may modify these associations, including age, sex, hospital visit type and season.

METHODS

Using a case-crossover design with distributed lag non-linear DTR terms (0-6 days), we estimated associations between ZIP Code-level DTR and hospital visits for mood (4.6 million hospital visits), anxiety (2.4 million), adjustment (∼368,000), and schizophrenia disorders (∼211,000), controlling for daily mean temperature, via conditional logistic regression models. We assessed potential heterogeneity by age, sex, hospital visit type (in-patient vs. out-patient), and season (summer, winter, and transition seasons).

RESULTS

For all included outcomes, we observed positive associations from period minimum DTR (0.1 °C) until 25th percentile (5.2 °C) and between mean DTR (7.7 °C) and 90th percentile (12.2 °C), beyond which we observed negative associations. For mood disorders, an increase in DTR from 0.1 °C to 12.2 °C was associated with a cumulative 16.0% increase (95% confidence interval [CI]: 12.8, 19.2%) in hospital visit rates. This increase was highest during transition seasons (32.5%; 95%CI: 26.4, 39.0%) compared with summer (10.7%; 95%CI: 4.8, 16.8%) and winter (-1.6%; 95%CI: -7.6, 4.7%). For adjustment and schizophrenia disorders, the strongest associations were seen among the youngest group (0-24 years) with almost no association in the oldest group (65+ years). We observed no evidence for modification by sex and hospital visit type.

DISCUSSION

Daily temperature variability was positively associated with mental health-related hospital visits within specific DTR ranges in New York State, after controlling for daily mean temperature. Given uncertainty on how climate change modifies temperature variability, additional research is crucial to comprehend the implications of these findings, particularly under different scenarios of future temperature variability.

Tattoos as a risk factor for malignant lymphoma: a population-based case-control study

Background

The popularity of tattoos has increased dramatically over the last few decades. Tattoo ink often contains carcinogenic chemicals, e.g., primary aromatic amines, polycyclic aromatic hydrocarbons, and metals. The tattooing process invokes an immunologic response that causes translocation of tattoo ink from the injection site. Deposition of tattoo pigment in lymph nodes has been confirmed but the long-term health effects remain unexplored. We used Swedish National Authority Registers with full population coverage to investigate the association between tattoo exposure and overall malignant lymphoma as well as lymphoma subtypes.

Methods

We performed a case-control study where we identified all incident cases of malignant lymphoma diagnosed between 2007 and 2017 in individuals aged 20-60 years in the Swedish National Cancer Register. Three random age- and sex-matched controls per case were sampled from the Total Population Register using incidence density sampling. We assessed exposure through a questionnaire in 2021, and data on potential confounders were retrieved from registers. We used multivariable logistic regression to estimate the incidence rate ratio (IRR) of malignant lymphoma in tattooed individuals.

Findings

The study population consisted of 11,905 individuals, and the response rate was 54% among cases (n = 1398) and 47% among controls (n = 4193). The tattoo prevalence was 21% among cases and 18% among controls. Tattooed individuals had a higher adjusted risk of overall lymphoma (IRR = 1.21; 95% CI 0.99-1.48). The risk of lymphoma was highest in individuals with less than two years between their first tattoo and the index year (IRR = 1.81; 95% CI 1.03-3.20). The risk decreased with intermediate exposure duration (three to ten years) but increased again in individuals who received their first tattoo ≥11 years before the index year (IRR = 1.19; 95% CI 0.94-1.50). We found no evidence of increasing risk with a larger area of total tattooed body surface. The risk associated with tattoo exposure seemed to be highest for diffuse large B-cell lymphoma (IRR 1.30; 95% CI 0.99-1.71) and follicular lymphoma (IRR 1.29; 95% CI 0.92-1.82).

Interpretation

Our findings suggested that tattoo exposure was associated with an increased risk of malignant lymphoma. More epidemiologic research is urgently needed to establish causality.

Funding

The Swedish Research Council for Health, Working Life and Welfare.

Can an Integrated Palliative and Oncology Co-rounding Model Reduce Aggressive Care at the End of Life? Secondary Analysis of an Open-label Stepped-wedge Cluster-randomized Trial

BACKGROUND

Clinical trial evidence on the effect of palliative care models in reducing aggressive end-of-life care is inconclusive. We previously reported on an integrated inpatient palliative care and medical oncology co-rounding model that significantly reduced hospital bed-days and postulate additional effect on reducing care aggressiveness.

OBJECTIVES

To compare the effect of a co-rounding model vs usual care in reducing receipt of aggressive treatment at end-of-life.

METHODS

Secondary analysis of an open-label stepped-wedge cluster-randomized trial comparing two integrated palliative care models within the inpatient oncology setting. The co-rounding model involved pooling specialist palliative care and oncology into one team with daily review of admission issues, while usual care constituted discretionary specialist palliative care referrals by the oncology team. We compared odds of receiving aggressive care at end-of-life: acute healthcare utilization in last 30 days of life, death in hospital, and cancer treatment in last 14 days of life between patients in two trial arms.

RESULTS

2145 patients were included in the analysis, and 1803 patients died by 4th April 2021. Median overall survival was 4.90 (4.07 - 5.72) months in co-rounding and 3.75 (3.22 - 4.21) months in usual care, with no difference in survival (P = .12). We found no significant differences between both models with respect to receipt of aggressive care at end-of-life. (Odds Ratio .67 - 1.27; all P > .05).

CONCLUSION

The co-rounding model within an inpatient setting did not reduce aggressiveness of care at end-of-life. This could be due in part to the overall focus on resolving episodic admission issues.

Effect of Smoking Cessation on the Risk of Subarachnoid Hemorrhage: A Nested Case-Control Study in Korean Men

BACKGROUND

Smoking is a well-established risk factor for subarachnoid hemorrhage (SAH), and current smokers have an increased risk of SAH. However, there are insufficient data on whether smoking cessation in current smokers reduces the risk of SAH.

METHODS

A nested case-control study was conducted based on the National Health Insurance Service-National Health Screening Cohort, which comprises nationwide health claims data and a national health screening program in Korea (2002-2019). We constructed a cohort of current male smokers without a history of stroke at the baseline health screening (2002-2003). From this cohort, cases were defined as individuals who experienced a nontraumatic SAH during the follow-up period up to 2019. Five controls were matched to each case using incidence density sampling. Smoking status (continuation or cessation) before the occurrence of SAH was evaluated using the repeated national health screening program. We conducted a multivariable conditional logistic regression analysis, adjusting for alcohol consumption, physical activity, body mass index, systolic blood pressure, and fasting blood glucose levels, to evaluate the association between SAH risk and smoking cessation.

RESULTS

At baseline, there were 112 142 current male smokers. After excluding individuals with prior stroke or insufficient data, the cohort consisted of 105 223 eligible participants (mean age, 50.3±8.5 years). Among them, we identified 318 cases of SAH and 1590 matched controls. Those who quit smoking had a significantly lower risk of SAH compared with current smokers (adjusted odds ratio, 0.55 [95% CI, 0.41-0.73]). The risk of SAH decreased with a longer period of smoking cessation. The risk reduction with smoking cessation was consistent even among prior heavy smokers.

CONCLUSIONS

Smoking cessation in current male smokers reduced the risk of SAH, and the risk reduction was greater as the cessation period increased. These findings warrant intensive efforts to encourage smokers, even heavy smokers, to quit.

PM2.5 exposure and risk of lung adenocarcinoma in women of Taiwan: A case-control study with density sampling

BACKGROUND AND OBJECTIVE

The prevalence of smoking among women in Taiwan is <5%, but the incidence of lung cancer remains high. This study determined the association between PM_2.5 (fine particulate matter with an aerodynamic diameter of ≤2.5 μm) exposure and lung cancer among women in Taiwan.

METHODS

In total, 21,301 female lung cancer cases nationwide were newly diagnosed between 2012 and 2017. Each case was age-, sex- and calendar year-matched with four controls randomly selected from the general population. Allowing a latent period of 5 years, we estimated the PM_2.5 and nitrogen dioxide (NO₂ ) exposures for each individual according to the residential changes from 2000. We adopted self-reported smoking statuses for the cases, while those of controls were estimated using annual surveys in each residential county. We performed multiple logistic regression analyses to examine the associations between PM_2.5 and NO₂ exposures and incident lung cancer cases.

RESULTS

The ORs of lung adenocarcinoma for the third (30.5-35.1 μg/m³ ), fourth (35.1-39.3 μg/m³ ) and fifth PM_2.5 exposure quintiles (39.3-48.1 μg/m³ ) relative to the first quintile were 1.10 (95% CI: 1.04-1.16), 1.12 (95% CI: 1.06-1.19) and 1.10 (95% CI: 1.04-1.16), respectively, after adjusting for smoking, residence and comorbidities. A dose-response relationship (p = 0.004) was found. The associations persisted with a 10-year latency and were not detected for small-cell and squamous cell carcinoma after control for smoking. We did not observe a similar effect for NO₂ exposure.

CONCLUSION

Residential PM_2.5 exposure higher than 30 μg/m³ was associated with an increased risk of lung adenocarcinoma in women of Taiwan.

Risk factors for hospital-acquired bacteraemia - an explorative case-control study of hospital interventions

BACKGROUND

Knowledge on hospital-related interventions as risk factors for hospital-acquired bacteraemia (HAB) is sparse.

AIM

We aimed to investigate hospital interventions as risk factors for HAB.

METHODS

Prospectively through one year, we identified episodes of HAB in a single tertiary hospital. We used a matched incidence density sampled case-control design. Matching on sex and age group, we sampled controls (1:2) from the adult hospital population with ongoing hospitalization for ≥48 h. Using conditional logistic regression, we estimated odds ratios (OR) with 95% confidence intervals (CI). For adjusted ORs (aOR), adjustments were made for length of hospital stay, type and urgency of admission, and Charlson Comorbidity Index score level.

FINDINGS

From 15th October 2019 through 14th October 2020, we identified 115 incident episodes of HAB and matched them with 230 controls. HAB patients were more often admitted as 'medicine or emergency surgery'-patients (94% vs 87%) and had a longer hospital stay before inclusion (median days 20 vs 12). They were more frequently categorized as having a 'low level comorbidity' (58% vs 39%) but had higher prevalence of haematologic (15% vs 6%) or metastatic cancer (13% vs 10%). Our estimates for central venous catheters were aOR of 3.46 (95% CI 1.92-6.23), haemodialysis; aOR 5.05 (95% CI 1.41-18.06), immunosuppressive treatment including chemotherapy; aOR of 1.72 (95% CI 1.00-2.96).

CONCLUSION

Central venous catheters and haemodialysis were the most prominent risk factors. Immunosuppressive treatment including therapy may play an important role in the development of HAB.

Random control selection for conducting high-throughput adverse drug events screening using large-scale longitudinal health data

Case-control design based high-throughput pharmacoinformatics study using large-scale longitudinal health data is able to detect new adverse drug event (ADEs) signals. Existing control selection approaches for case-control design included the dynamic/super control selection approach. The dynamic/super control selection approach requires all individuals to be evaluated at all ADE case index dates, as the individuals' eligibilities as control depend on ADE/enrollment history. Thus, using large-scale longitudinal health data, the dynamic/super control selection approach requires extraordinarily high computational time. We proposed a random control selection approach in which ADE case index dates were matched by randomly generated control index dates. The random control selection approach does not depend on ADE/enrollment history. It is able to significantly reduce computational time to prepare case-control data sets, as it requires all individuals to be evaluated only once. We compared the performance metrics of all control selection approaches using two large-scale longitudinal health data and a drug-ADE gold standard including 399 drug-ADE pairs. The F-scores for the random control selection approach were between 0.586 and 0.600 compared to between 0.545 and 0.562 for dynamic/super control selection approaches. The random control selection approach was ~ 1000 times faster than dynamic/super control selection approach on preparing case-control data sets. With large-scale longitudinal health data, a case-control design-based pharmacoinformatics study using random control selection is able to generate comparable ADE signals than the existing control selection approaches. The random control selection approach also significantly reduces computational time to prepare the case-control data sets.

Matched Versus Unmatched Analysis of Matched Case-Control Studies

Although the need for addressing matching in the analysis of matched case-control studies is well established, debate remains as to the most appropriate analytical method when matching on at least 1 continuous factor. We compared the bias and efficiency of unadjusted and adjusted conditional logistic regression (CLR) and unconditional logistic regression (ULR) in the setting of both exact and nonexact matching. To demonstrate that case-control matching distorts the association between the matching variables and the outcome in the matched sample relative to the target population, we derived the logit model for the matched case-control sample under exact matching. We conducted simulations to validate our theoretical conclusions and to explore different ways of adjusting for the matching variables in CLR and ULR to reduce biases. When matching is exact, CLR is unbiased in all settings. When matching is not exact, unadjusted CLR tends to be biased, and this bias increases with increasing matching caliper size. Spline smoothing of the matching variables in CLR can alleviate biases. Regardless of exact or nonexact matching, adjusted ULR is generally biased unless the functional form of the matched factors is modeled correctly. The validity of adjusted ULR is vulnerable to model specification error. CLR should remain the primary analytical approach.

Decreased Susceptibility of Marginal Odds Ratios to Finite-sample Bias

Parameters representing adjusted treatment effects may be defined marginally or conditionally on covariates. The choice between a marginal or covariate-conditional parameter should be driven by the study question. However, an unappreciated benefit of marginal estimators is a reduction in susceptibility to finite-sample bias relative to the unpenalized maximum likelihood estimator of the covariate-conditional odds ratio (OR). Using simulation, we compare the finite-sample bias of different marginal and conditional estimators of the OR. We simulated a logistic model to have 15 events per parameter and two events per parameter. We estimated the covariate-conditional OR by maximum likelihood with and without Firth's penalization. We used three estimators of the marginal OR: g-computation, inverse probability of treatment weighting, and augmented inverse probability of treatment weighting. At 15 events per parameter, as expected, all estimators were effectively unbiased. At two events per parameter, the unpenalized covariate-conditional estimator was notably biased but penalized covariate-conditional and marginal estimators exhibited minimal bias.

Association between allergic or autoimmune diseases and incidence of endometriosis: A nested case-control study using a health insurance claims database

PROBLEM

Immune responses were reported to be associated with the pathogenesis of endometriosis. However, previous studies of an association between allergic or autoimmune diseases and endometriosis have reported inconsistent results. We investigated the association between allergic or autoimmune diseases and the incidence of endometriosis.

METHOD OF STUDY

Using a large-scale health insurance claims database in Japan, we identified patients with endometriosis diagnosed between April 2011 and August 2018. For each case, we identified up to four controls with the same age and registration month. Conditional logistic regression analyses were conducted to evaluate the incidence rate ratio (IRR) of endometriosis in women with allergic diseases (asthma, allergic rhinitis, urticaria, atopic dermatitis, allergic conjunctivitis, and type 1 allergies combined), systemic lupus erythematosus, and rheumatoid arthritis compared with women without these diseases.

RESULTS

We identified 30 516 cases with endometriosis and 120 976 control participants. Mean age at registration was 30 years. There were significant positive associations between type 1 allergy and endometriosis (IRR, 1.10; 95% confidence interval, 1.06-1.13) and between rheumatoid arthritis and endometriosis (IRR, 1.31; 95% confidence interval, 1.05-1.64). Diagnosis of systemic lupus erythematosus was not associated with increased incidence of endometriosis. Among the different allergic diseases, allergic rhinitis, urticaria, and allergic conjunctivitis were associated with the increased incidence.

CONCLUSION

Several allergic diseases were associated with an increased incidence of endometriosis. A higher incidence was also observed in patients with rheumatoid arthritis. Further studies are warranted to elucidate the influence of immune responses on the development of endometriosis.

Statistical inference in matched case-control studies of recurrent events

BACKGROUND

The concurrent sampling design was developed for case-control studies of recurrent events. It involves matching for time. Standard conditional logistic-regression (CLR) analysis ignores the dependence among recurrent events. Existing methods for clustered observations for CLR do not fit the complex data structure arising from the concurrent sampling design.

METHODS

We propose to break the matches, apply unconditional logistic regression with adjustment for time in quintiles and residual time within each quintile, and use a robust standard error for observations clustered within persons. We conducted extensive simulation to evaluate this approach and compared it with methods based on CLR. We analysed data from a study of childhood pneumonia to illustrate the methods.

RESULTS

The proposed method and CLR methods gave very similar point estimates of association and showed little bias. The proposed method produced confidence intervals that achieved the target level of coverage probability, whereas the CLR methods did not, except when disease incidence was low.

CONCLUSIONS

The proposed method is suitable for the analysis of case-control studies with recurrent events.

Association Between Alzheimer Disease and Cancer With Evaluation of Study Biases: A Systematic Review and Meta-analysis

IMPORTANCE

Observational studies consistently report inverse associations between cancer and Alzheimer disease (AD). Shared inverse etiological mechanisms might explain this phenomenon, but a systematic evaluation of methodological biases in existing studies is needed.

OBJECTIVES

To systematically review and meta-analyze evidence on the association between cancer and subsequent AD, systematically identify potential methodological biases in studies, and estimate the influence of these biases on the estimated pooled association between cancer and AD.

DATA SOURCES

All-language publications were identified from PubMed, Embase, and PsycINFO databases through September 2, 2020.

STUDY SELECTION

Longitudinal cohort studies and case-control studies on the risk of AD in older adults with a history of any cancer type, prostate cancer, breast cancer, colorectal cancer, or nonmelanoma skin cancer, relative to those with no cancer history.

DATA EXTRACTION AND SYNTHESIS

Two reviewers independently abstracted the data and evaluated study biases related to confounding, diagnostic bias, competing risks, or survival bias. Random-effects meta-analysis was used to provide pooled estimates of the association between cancer and AD. Metaregressions were used to evaluate whether the observed pooled estimate could be attributable to each bias. The study was designed and conducted according to the Preferring Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline.

MAIN OUTCOMES AND MEASURES

Incidence, hazard, or odds ratios for AD comparing older adults with vs without a previous cancer diagnosis.

RESULTS

In total, 19 cohort studies and 3 case-control studies of the associations between any cancer type (n = 13), prostate cancer (n = 5), breast cancer (n = 1), and nonmelanoma skin cancer (n = 3) with AD were identified, representing 9 630 435 individuals. In all studies combined, cancer was associated with decreased AD incidence (cohort studies: random-effects hazard ratio, 0.89; 95% CI, 0.79-1.00; case-control studies: random-effects odds ratio, 0.75; 95% CI, 0.61-0.93). Studies with insufficient or inappropriate confounder control or greater likelihood of AD diagnostic bias had mean hazard ratios closer to the null value, indicating that these biases could not explain the observed inverse association. Competing risks bias was rare. Studies with greater likelihood of survival bias had mean hazard ratios farther from the null value.

CONCLUSIONS AND RELEVANCE

The weak inverse association between cancer and AD may reflect shared inverse etiological mechanisms or survival bias but is not likely attributable to diagnostic bias, competing risks bias, or insufficient or inappropriate control for potential confounding factors.