Search

ObjectiveIndividuals with ankylosing spondylitis (AS) have a greater cardiovascular (CV) risk than those in the general population. The effect of tumor necrosis factor inhibitors (TNFis) on CV risk, including on the development of hypertension (HTN), remains unclear, with some data suggesting higher risk. We assessed the association of TNFi use with incident HTN in a longitudinal AS cohort.MethodsAdults with AS enrolled in a prospective cohort in 2002–2018 were examined every 4–6 months. TNFi use during the preceding 6 months was ascertained at each study visit. We defined HTN by patient-reported HTN, antihypertensive medication use, or, on 2 consecutive visits, systolic blood pressure (BP) ≥ 140 mmHg or diastolic BP ≥ 90 mmHg. We evaluated the association between TNFi use and the development of HTN with marginal structural models, estimated by inverse probability-of-treatment weighting, to account for time-dependent confounders and informative censoring. Potential confounders included age, sex, race, site, nonsteroidal antiinflammatory drug use, and disease activity.ResultsWe included 630 patients without baseline HTN and with at least 1 year of follow-up. Of these, 72% were male, mean age was 39 ± 13 years, and 43% used TNFi at baseline. On follow-up (median 5 yrs), 129 developed incident HTN and 163 started on TNFi during follow-up. TNFi use was not associated with incident HTN (adjusted HR 1.10, 95% CI 0.83–1.37).ConclusionIn our prospective AS cohort, TNFi use was not significantly associated with incident HTN.

Objective Hypomagnesemia increases the risk of chondrocalcinosis and calcium pyrophosphate disease. We examined whether the use of drugs that can cause hypomagnesemia, diuretics and proton pump inhibitors (PPIs), increases the risk of chondrocalcinosis. Methods Participants in the Multicenter Osteoarthritis (MOST) Study obtained weight‐bearing knee radiographs, and their medication use was recorded at baseline and 30‐, 60‐, 84‐, and 144‐month examinations. We read radiographs serially for chondrocalcinosis and characterized incident chondrocalcinosis when it first appeared. We classified diuretic use as thiazide, loop, and other. To test drug effects on incident chondrocalcinosis at each interval (eg, 30‐60 months), we excluded persons with chondrocalcinosis at the interval’s beginning. For each drug, we evaluated exposure at the beginning and end of the interval. We conducted knee‐based analyses using Bayesian mixed‐effects discrete time survival models adjusted for age, sex, body mass index, radiographic osteoarthritis, race, and clinic site. Results Of 5272 knees, 196 developed chondrocalcinosis. Thiazide use (21.7% of examinations) and PPI use (13.7%) were common. Neither loop nor other diuretic use was associated with incident chondrocalcinosis. Thiazide use at the beginning and end of the interval of incidence conferred a high risk (hazard ratio [HR] = 2.18; 95% confidence interval [CI] 1.23‐3.89), but use at the beginning of the interval was not associated with risk (HR = 1.04). PPI use at the interval’s beginning increased risk of chondrocalcinosis (HR = 2.29; 95% CI 1.37‐3.79). Conclusion Thiazide diuretics, but not other diuretics, and PPI use probably increase the risk of chondrocalcinosis. These findings may have important clinical implications.

Gait alterations in those with mild unilateral knee pain during walking may provide clues to modifiable alterations that affect progression of knee pain and osteoarthritis(OA). To examine this, we applied machine learning (ML) approaches to gait data from wearable sensors in a large observational knee OA cohort, the Multicenter Osteoarthritis (MOST) study. Participants completed a 20-meter walk test wearing sensors on their trunk and ankles. Parameters describing spatiotemporal features of gait and symmetry, variability and complexity were extracted. We used an ensemble ML technique (“super learning”) to identify gait variables in our cross-sectional data associated with the presence/absence of unilateral knee pain. We then used logistic regression to determine the association of selected gait variables with odds of mild knee pain.Of 2066 participants (mean age 63.6 [SD: 10.4] years, 56% female), 21.3% had mild unilateral pain while walking. Gait parameters selected in the ML process as influential included step regularity, sample entropy, gait speed, and amplitude dominant frequency, among others. In adjusted cross-sectional analyses, lower levels of step regularity (i.e., greater gait variability) and lower sample entropy(i.e., lower gait complexity) were associated with increased likelihood of unilateral mild pain while walking [aOR=0.80 (0.64-1.00) and aOR=0.79(0.66-0.95), respectively].

ObjectivesAlthough treatment development in osteoarthritis (OA) focuses on chondroprotection, it is unclear how much preventing cartilage loss reduces joint pain. It is also unclear how nociceptive tissues may be involved.MethodsUsing data from the Osteoarthritis Initiative, we quantified the relation between cartilage loss and worsening knee pain after adjusting for bone marrow lesions (BMLs) and synovitis, and examined how much these factors mediated this association. 600 knee MRIs were scored at baseline, 12 months and 24 months for quantitative and semiquantitative measures of OA structural features. We focused on change in medial cartilage thickness using an amount similar to that seen in recent trials. Linear models calculated mean change in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain score with cartilage loss, adjusted for baseline BMLs, synovitis and covariates. Mediation analysis tested whether change in synovitis or BMLs mediated the cartilage loss–pain association. We carried out a subanalysis for knees with non-zero baseline WOMAC pain scores and another for non-valgus knees.ResultsCartilage thickness loss was significantly associated with a small degree of worsening in pain over 24 months. For example, a loss of 0.1 mm of cartilage thickness over 2 years was associated with a 0.32 increase in WOMAC pain (scale 0–20). The association of cartilage thickness loss with pain was mediated by synovitis change but not by BML change. Subanalysis results were similar.ConclusionsCartilage thickness loss is associated with only a small amount of worsening knee pain, an association mediated in part by worsening synovitis. Demonstrating that chondroprotection reduces knee pain will be extremely challenging and is perhaps unachievable.

Knee replacement (KR) rates are increasing exponentially in the US and straining insurance budgets. This study was undertaken to investigate how many KRs would be prevented at different levels of pain improvement, a major target of osteoarthritis (OA) trials.We used data from the Osteoarthritis Initiative (OAI) to emulate a trial of knee pain interventions on KR risk changes. We modeled hypothetical 1-, 2- or 3-unit reductions of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain subscale whenever a person reported a pain score of ≥5 (of 20) in an affected knee at any clinic visit. We used causal inference-based targeted learning to estimate treatment effects for hypothesized pain intervention strategies adjusted for time-dependent confounding. Sensitivity analyses assessed interventions at WOMAC pain scores of ≥4 and ≥7.Of the 9,592 knees studied (n = 4,796 participants; 58.5% female; baseline age 61.2 years), 40.7% experienced WOMAC pain scores of ≥5. The estimated knee-level (reference) risk of a KR, adjusted for loss to follow-up and death, was 6.3% (95% confidence interval 5.0, 7.7%) in the OAI. Reductions of WOMAC pain scores by 1, 2, or 3 units decreased the KR risk from 6.3% to 5.8%, 5.3%, and 4.9%, respectively. Larger reductions in KR risk were achieved when interventions were applied at a WOMAC pain score of ≥4.Modest pain reductions from OA interventions would substantially reduce the number of KRs, with greater reductions achieved when pain decreased more and when interventions were introduced at lower pain levels.

Objective To examine the association of body mass index (BMI) with pain in people with hand osteoarthritis (OA), and explore whether this association, if causal, is mediated by systemic inflammatory biomarkers. Methods In 281 Nor-Hand study participants, we estimated associations between BMI and hand pain, as measured by the Australian/Canadian Osteoarthritis Hand Index (AUSCAN; range 0–20) and Numerical Rating Scale (NRS; range 0–10); foot pain, as measured by NRS (range 0–10); knee/hip pain, as measured by the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC; range 0–20); painful total body joint count; and pain sensitization. We fit natural-effects models to estimate natural direct and natural indirect effects of BMI on pain through inflammatory biomarkers. Results Each 5-unit increase in BMI was associated with more severe hand pain (on average increased AUSCAN by 0.64 [95% confidence interval (95% CI) 0.23, 1.08]), foot pain (on average increased NRS by 0.65 [95% CI 0.36, 0.92]), knee/hip pain (on average increased WOMAC by 1.31 [95% CI 0.87, 1.73]), generalized pain, and pain sensitization. Mediation analyses suggested that the effects of BMI on hand pain and painful total body joint count were partially mediated by leptin and high-sensitivity C-reactive protein (hsCRP), respectively. Effect sizes for mediation by leptin were larger for the hands than for the lower extremities, and were statistically significant for the hands only. Conclusion In people with hand OA, higher BMI is associated with greater pain severity in the hands, feet, and knees/hips. Systemic effects of obesity, measured by leptin, may play a larger mediating role for pain in the hands than in the lower extremities. Low-grade inflammation, measured by hsCRP, may contribute to generalized pain in overweight/obese individuals.

ImportancePatients with septic shock may benefit from the initiation of corticosteroids. However, the comparative effectiveness of the 2 most studied corticosteroid regimens (hydrocortisone with fludrocortisone vs hydrocortisone alone) is unclear.ObjectiveTo compare the effectiveness of adding fludrocortisone to hydrocortisone vs hydrocortisone alone among patients with septic shock using target trial emulation.Design, Setting, and ParticipantsThis retrospective cohort study from 2016 to 2020 used the enhanced claims-based Premier Healthcare Database, which included approximately 25% of US hospitalizations. Participants were adult patients hospitalized with septic shock and receiving norepinephrine who began hydrocortisone treatment. Data analysis was performed from May 2022 to December 2022.ExposureAddition of fludrocortisone on the same calendar day that hydrocortisone treatment was initiated vs use of hydrocortisone alone.Main Outcome and MeasuresComposite of hospital death or discharge to hospice. Adjusted risk differences were calculated using doubly robust targeted maximum likelihood estimation.ResultsAnalyses included 88 275 patients, 2280 who began treatment with hydrocortisone-fludrocortisone (median [IQR] age, 64 [54-73] years; 1041 female; 1239 male) and 85 995 (median [IQR] age, 67 [57-76] years; 42 136 female; 43 859 male) who began treatment with hydrocortisone alone. The primary composite outcome of death in hospital or discharge to hospice occurred among 1076 (47.2%) patients treated with hydrocortisone-fludrocortisone vs 43 669 (50.8%) treated with hydrocortisone alone (adjusted absolute risk difference, −3.7%; 95% CI, −4.2% to −3.1%; P Conclusions and RelevanceIn this comparative effectiveness cohort study among adult patients with septic shock who began hydrocortisone treatment, the addition of fludrocortisone was superior to hydrocortisone alone.

ObjectiveTo 1) develop and evaluate a machine learning model incorporating gait and physical activity to predict medial tibiofemoral cartilage worsening over two years in individuals without or with early knee osteoarthritis and 2) identify influential predictors in the model and quantify their effect on cartilage worsening.DesignAn ensemble machine learning model was developed to predict worsened cartilage MRI Osteoarthritis Knee Score at follow-up from gait, physical activity, clinical and demographic data from the Multicenter Osteoarthritis Study. Model performance was evaluated in repeated cross-validations. The top 10 predictors of the outcome across 100 held-out test sets were identified by a variable importance measure statistic, and their marginal effect on the outcome was quantified by g-computation.ResultsOf 947 legs in the analysis, 14% experienced medial cartilage worsening at follow-up. The median (2.5th-97.5thpercentile) AUC across the 100 held-out test sets was 0.73 (0.65-0.79). Presence of baseline cartilage damage, higher Kellgren-Lawrence grade, greater pain during walking, higher lateral ground reaction force impulse, greater time spent lying, and lower vertical ground reaction force unloading rate were associated with greater risk of cartilage worsening.ConclusionsAn ensemble machine learning approach incorporating gait, physical activity, and clinical/demographic features showed good performance for predicting cartilage worsening over two years. While identifying potential intervention targets from the model is challenging, these results suggest that lateral ground reaction force impulse, time spent lying, and vertical ground reaction force unloading rate should be investigated further as potential early intervention targets to reduce medial tibiofemoral cartilage worsening.Summary boxWhat are the findings?Machine learning models predicted cartilage worsening in persons without or with early knee osteoarthritis from gait, physical activity, and clinical and demographic characteristics with a median AUC of 0.73 across 100 held-out test sets.High lateral ground reaction force impulse, more time spent lying, and low vertical ground reaction force unloading rate were associated with increased risk of cartilage worsening over two years.How might it impact on clinical practice in the future?Gait and physical activity are some of the only modifiable risk factors for knee osteoarthritis; this study identified three potential intervention targets to slow early knee osteoarthritis progression.

OBJECTIVE Massive weight loss leads to marked knee pain reduction in individuals with knee pain, but the reason for the reduction in pain is unknown. This study was undertaken to quantify the contribution of magnetic resonance imaging (MRI)-evidenced changes in pain-sensitive structures, bone marrow lesions (BMLs), and synovitis, and changes in pain sensitization or depressive symptoms, to knee pain improvement after substantial weight loss. METHODS Morbidly obese patients with knee pain on most days were evaluated before bariatric surgery or medical weight management and at 1-year follow-up for BMLs and synovitis seen on MRI, the pressure pain threshold (PPT) at the patella and the right wrist, depressive symptoms (using the Center for Epidemiologic Studies Depression scale [CES-D]), and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain survey. Natural-effects models were used to quantify the extent that achieving a minimum clinically important difference (MCID) of ≥18% on the WOMAC pain scale could be mediated by weight loss-induced changes in BMLs, synovitis, PPT, and depressive symptoms. RESULTS Of 75 participants, 53.3% lost ≥20% of weight by 1 year. Of these, 75% attained the MCID for pain improvement, compared with 34.3% in those who had

OBJECTIVE To evaluate the association of fatty infiltration of the quadriceps and vastus medialis (VM) with an increase in knee cartilage, meniscus, or bone marrow lesions, using magnetic resonance imaging (MRI) in knee osteoarthritis (OA) over 3 years. METHODS Participants (n = 69) with and without radiographic knee OA underwent MRI at baseline and 3 years later. Chemical shift-based water/fat MRI was used to quantify the intramuscular fat fraction and the lean anatomical cross-sectional area (ACSA) for the VM and entire quadriceps muscles. MRI images of the knee were analyzed using the semiquantitative modified whole-organ MRI score (mWORMS) grading to assess change in lesions in the articular cartilage, meniscus, and bone marrow. Logistic regression was used to assess whether baseline quadriceps and VM fat fraction and lean ACSA were associated with an increase in mWORMS scores. Odds ratios (ORs) were adjusted for age, sex, and body mass index. RESULTS Overall, of the 69 subjects, 43 (62%) had an increase in cartilage lesions (26 of 43), meniscus lesions (19 of 43), or bone marrow lesions (22 of 43) scores. The quadriceps (OR 2.13 [95% confidence interval (95% CI) 1.09-4.15]) and VM (OR 2.05 [95% CI 1.25-3.36]) fat fraction were both associated with an increase in cartilage, meniscus, or bone marrow lesion scores over 3 years. The association of quadriceps or VM lean ACSA with the outcomes was not significant. CONCLUSION These longitudinal findings using quantitative MRI methods for assessment of muscle adiposity highlight the role of quadriceps adiposity, specifically in the VM, in knee OA progression. However, studies in larger cohorts are needed to confirm these findings.

Objective: The lack of strong association between knee osteoarthritis (OA) structural features and pain continues to perplex researchers and clinicians. Evaluating the patellofemoral joint in addition to the tibiofemoral joint alone has contributed to explaining this structure–pain discordance, hence justifying a more comprehensive evaluation of whole-knee OA and pain. The present study, therefore, was undertaken to evaluate the association between patellofemoral and tibiofemoral OA features with localized anterior knee pain (AKP) using 2 study designs. Methods: Using cross-sectional data from the Multicenter Osteoarthritis Study, our first approach was a within-person, knee-matched design in which we identified participants with unilateral AKP. We then assessed magnetic resonance imaging (MRI)–derived OA features (cartilage damage, bone marrow lesions [BMLs], osteophytes, and inflammation) in both knees and evaluated the association of patellofemoral and tibiofemoral OA features to unilateral AKP. In our second approach, MRIs from 1 knee per person were scored, and we evaluated the association of OA features to AKP in participants with AKP and participants with no frequent knee pain. Results: Using the first approach (n = 71, 66% women, mean ± SD age 69 ± 8 years), lateral patellofemoral osteophytes (odds ratio [OR] 5.0 [95% confidence interval (95% CI) 1.7–14.6]), whole-knee joint effusion-synovitis (OR 4.7 [95% CI 1.3–16.2]), and infrapatellar synovitis (OR 2.8 [95% CI 1.0–7.8]) were associated with AKP. Using the second approach (n = 882, 59% women, mean ± SD age 69 ± 7 years), lateral and medial patellofemoral cartilage damage (prevalence ratio [PR] 2.3 [95% CI 1.3–4.0] and PR 1.9 [95% CI 1.1–3.3], respectively) and lateral patellofemoral BMLs (PR 2.6 [95% CI 1.5–4.7]) were associated with AKP. Conclusion: Patellofemoral but not tibiofemoral joint OA features and inflammation were associated with AKP.

Background The effectiveness of interleukin-6 inhibitors (IL-6i) in ameliorating coronavirus disease 2019 (COVID-19) remains uncertain. Methods We analyzed data for patients aged ≥18 years admitted with a positive severe acute respiratory syndrome coronavirus 2 polymerase chain reaction test at 4 safety-net hospital systems with diverse populations and high rates of medical comorbidities in 3 US regions. We used inverse probability of treatment weighting via machine learning for confounding adjustment by demographics, comorbidities, and disease severity markers. We estimated the average treatment effect, the odds of IL-6i effect on in-hospital mortality from COVID-19, using a logistic marginal structural model. Results Of 516 patients, 104 (20.1%) received IL-6i. Estimate of the average treatment effect adjusted for confounders suggested a 37% reduction in odds of in-hospital mortality in those who received IL-6i compared with those who did not, although the confidence interval included the null value of 1 (odds ratio = 0.63; 95% confidence interval, .29–1.38). A sensitivity analysis suggested that potential unmeasured confounding would require a minimum odds ratio of 2.55 to nullify our estimated IL-6i effect size. Conclusions Despite low precision, our findings suggested a relatively large effect size of IL-6i in reducing the odds of COVID-19–related in-hospital mortality.

Administrative health databases have been used to monitor trends in infective endocarditis hospitalization related to nonprescription injection drug use (IDU) using International Classification of Diseases (ICD) code algorithms. Because no ICD code for IDU exists, drug dependence and hepatitis C virus (HCV) have been used as surrogate measures for IDU, making misclassification error (ME) a threat to the accuracy of existing estimates. In a serial cross-sectional analysis, we compared the unadjusted and ME-adjusted prevalences of IDU among 70,899 unweighted endocarditis hospitalizations in the 2007–2016 National Inpatient Sample. The unadjusted prevalence of IDU was estimated with a drug algorithm, an HCV algorithm, and a combination algorithm (drug and HCV). Bayesian latent class models were used to estimate the median IDU prevalence and 95% Bayesian credible intervals and ICD algorithm sensitivity and specificity. Sex- and age group-stratified IDU prevalences were also estimated. Compared with the misclassification-adjusted prevalence, unadjusted estimates were lower using the drug algorithm and higher using the combination algorithm. The median ME-adjusted IDU prevalence increased from 9.7% (95% Bayesian credible interval (BCI): 6.3, 14.8) in 2008 to 32.5% (95% BCI: 26.5, 38.2) in 2016. Among persons aged 18–34 years, IDU prevalence was higher in females than in males. ME adjustment in ICD-based studies of injection-related endocarditis is recommended.

BACKGROUND: Dietary fiber reduces body weight and inflammation in clinical trials. It is unclear whether body mass index (BMI) and inflammation might explain the observed association between higher fiber intake and the lower risk of symptomatic knee osteoarthritis (SXKOA). OBJECTIVES: We quantified the extent to which BMI and inflammation influenced the relation between dietary fiber and SXKOA. METHODS: We used longitudinal data from the Osteoarthritis Initiative (OAI) and the Framingham Offspring Osteoarthritis Study. At baseline of each study, men and women (mean age: 61 y) with or at risk of knee osteoarthritis were followed for 48 mo in the OAI. Adults (mean age: 53 y) were followed for 9.5 y in the Framingham study. Dietary fiber intake was estimated using a validated food-frequency questionnaire. Measured weight and height were used to calculate BMI. Serum high-sensitivity C-reactive protein (CRP) was measured in the Framingham study only. Incident SXKOA was defined as new onset of a combination of knee pain and radiographic osteoarthritis. We applied marginal structural models to quantify the mediation through BMI in the OAI and the sequential mediation through BMI and CRP in the Framingham study. RESULTS: Incident SXKOA occurred in 861 knees among 2876 persons in the OAI and in 143 knees among 971 persons in the Framingham study. In persons whose fiber intake was ≥21 g/d compared with those with intakes

The epidemiology of hospitalizations with infections among patients with rheumatoid arthritis (RA) is unknown, despite an increase in RA treatments that confer a risk of infection.We examined National Inpatient Sample data from 1993-2013. We identified hospitalizations among adults with RA, defined by International Classification of Diseases, Ninth Revision, Clinical Modification codes (714.xx) in any secondary diagnosis field. We evaluated 5 infections as the primary diagnosis: pneumonia, sepsis, urinary tract infection (UTI), skin and soft tissue infections (SSTIs), and opportunistic infections (OIs). The primary outcome was the proportion of hospitalizations for each infection among all hospitalizations with a secondary diagnosis of RA.There were 792,921 hospitalizations for infection with a secondary diagnosis of RA, with the rates increasing from 90 to 206 per 100,000 persons from 1993-2013. The proportion of hospitalizations decreased for pneumonia (from 5.4% to 4.6%), UTI (from 0.4% to 0.38%), and OIs (from 0.44% to 0.26%). The proportion of hospitalizations for SSTIs increased slightly (from 2.3% to 2.5%), while hospitalizations for sepsis more than tripled (from 1.9% to 6.4%).Between 1993 and 2013, the proportion of hospitalizations for infections among RA patients appeared to decline for pneumonia and OIs, with a slight decrease in UTI, a slight increase in SSTIs, and a substantial increase in hospitalizations with sepsis. Our results are consistent with previous reports that the sensitivity of sepsis coding has increased over time.

BackgroundUrinalysis and urine culture are commonly ordered tests in the emergency department (ED). We evaluated the impact of removal of order sets from the ‘frequently ordered test’ in the computerised physician order entry system (CPOE) on urine testing practices.MethodsWe conducted a before (1 September to 20 October 2015) and after (21 October to 30 November 2015) study of ED patients. The intervention consisted of retaining ‘urinalysis with reflex to microscopy’ as the only urine test in a highly accessible list of frequently ordered tests in the CPOE system. All other urine tests required use of additional order screens via additional mouse clicks. The frequency of urine testing before and after the intervention was compared, adjusting for temporal trends.ResultsDuring the study period, 6499 (28.2%) of 22 948 ED patients had ≥1 urine test ordered. Urine testing rates for all ED patients decreased in the post intervention period for urinalysis (291.5 pre intervention vs 278.4 per 1000 ED visits post intervention, P=0.03), urine microscopy (196.5vs179.5, P=0.001) and urine culture (54.3vs29.7, PConclusionsA simple intervention of retaining only ‘urinalysis with reflex to microscopy’ and removing all other urine tests from the ‘frequently ordered’ window of the ED electronic order set decreased urine cultures ordered by 46.6% after accounting for temporal trends. Given the injudicious use of antimicrobial therapy for asymptomatic bacteriuria, findings from our study suggest that proper design of electronic order sets plays a vital role in reducing excessive ordering of urine cultures.

Objective While salivary gland ultrasound (SGU) has gained prominence for evaluating Sjogren's syndrome, little information exists on SGU appearance of sarcoidosis and amyloidosis, potential mimics of Sjogren's syndrome. Our goal in this study was to estimate the diagnostic accuracy of major SGU features in differentiating Sjogren's syndrome from sarcoidosis, amyloidosis, and controls. Methods We enrolled consecutive adult ambulatory patients with a clinical diagnosis of Sjogren's syndrome fulfilling the 2016 American College of Rheumatology (ACR) classification criteria; we also enrolled patients with a clinical diagnosis of sarcoidosis or systemic immunoglobulin light chain (AL) amyloidosis, with histologic confirmation from any tissue, and rheumatology outpatients without diagnoses affecting salivary glands. Subjects underwent major SGU using the Hocevar protocol, with resulting video clips reviewed blind to clinical diagnosis. Results Sjogren's syndrome SGU scores were greater than in patients from the other groups, but there were no distinguishing salivary gland features from AL amyloidosis or sarcoidosis. None of the patients in the control group scored higher than 17, a cutoff previously suggested for Sjogren's syndrome, but 27% of patients with AL amyloidosis and 19% with sarcoidosis scored higher than 17. Adding Hocevar SGU scores of ≥17 to the 2016 ACR/European League Against Rheumatism criteria in a parallel scheme increased the sensitivity for Sjogren's syndrome from 87% to 98%, while combining the 2 criteria in series increased specificity from 81% to 98%. Conclusion Sjogren's syndrome, sarcoidosis, and AL amyloidosis share common SGU features that can help distinguish these conditions from patients without systemic rheumatologic disease. Clinicians should carefully consider these potential mimics when interpreting salivary gland US results.

Aim. To characterize antibiotic (ab) prescriptions in children. Methods. Evaluation of outpatient ab prescriptions in a 3-year cohort of children in primary care using a data warehouse (Massachusetts Health Disparities Repository) by comorbid conditions, demographics, and clinical indication. Results. A total of 15 208 children with nearly 120 000 outpatient visits were included. About one third had a comorbid condition (most commonly asthma). Among the 30 000 ab prescriptions, first-line penicillins and macrolides represented the most frequent ab (70%), followed by cephalosporins (16%). Comorbid children had 54.3 ab prescriptions/100 child-years versus 38.8 in children without comorbidity; ab prescription was higher in urinary tract infections (>60% of episodes), otitis, lower respiratory tract infections (>50%), especially in comorbid children and children under 2 year old. Ab prescriptions were significantly associated with younger age, emergency room visit, comorbid children, and acute infections. Discussion. A clinical data warehouse could help in designing appropriate antimicrobial stewardship programs and represent a potential assessment tool.

We present the first description of an antimicrobial stewardship program (ASP) used to successfully manage a multi-antimicrobial drug shortage. Without resorting to formulary restriction, meropenem utilization decreased by 69% and piperacillin-tazobactam by 73%. During the shortage period, hospital mortality decreased (P=.03), while hospital length of stay remained unchanged.Infect Control Hosp Epidemiol 2017;38:356–359

Our paper1 was primarily focused on the small effect that preventing cartilage loss has on pain reduction. In an additional analysis, we noted that synovitis change mediated a small percentage of that effect, an effect that was not statistically significant. Cashin et al 2 suggested that our mediation results should be interpreted based on whether the 95% CIs for the indirect or mediation effects overlapped with the null (ie, was statistically significant). Like the American Statistical Association and leading methodology experts, we oppose drawing a binary conclusion based on whether a threshold is statistically significant. Crossing an arbitrary significance threshold does not convey the presence or absence of a ‘real’ effect. …

Background International Classification of Diseases (ICD) code algorithms are routinely used to estimate the frequency of illicit injection drug use (IDU)-associated hospitalizations in administrative health datasets despite a lack of evidence regarding their validity. We aimed to measure the sensitivity and specificity of ICD code algorithms used to estimate the prevalence of current/recent IDU among infective endocarditis (IE) hospitalizations without a reference standard. Methods We reviewed medical records of 321 patients aged 18–64 years old from an urban academic hospital with an IE diagnosis between 2007 and 2017. Diagnostic tests for IDU included self-reported IDU in medical records; a drug use, abuse and dependence (UAD) ICD algorithm; a Hepatitis C Virus (HCV) ICD algorithm; and a combination drug UAD/HCV ICD algorithm. Sensitivity, specificity and the misclassification error (ME)-adjusted IDU prevalence were estimated using Bayesian latent class models. Results The combination algorithm had the highest sensitivity and lowest specificity. Sensitivity increased for the drug UAD algorithm in the ICD-10 period compared to the ICD-9 period. The ME-adjusted current/recent IDU prevalence estimated using the drug UAD and HCV algorithms was 23 % (95 % Bayesian credible interval: 16 %, 31 %). The unadjusted prevalence estimate from the drug UAD algorithm underestimated the ME-adjusted prevalence, while the combination algorithm overestimated it. Conclusion The validity of ICD code algorithms for IDU among IE hospitalizations is imperfect and differs between ICD-9 and ICD-10. Commonly used ICD-based algorithms could lead to substantially biased prevalence estimates in IDU-associated hospitalizations when using administrative health data.

Well-done randomized trials provide accurate estimates of treatment effect by producing groups that are similar on all measures except for the intervention of interest. However, inferences of efficacy in tightly-controlled experimental settings may not translate into similar effectiveness in real-world settings. Observational studies generally enable inferences over a wider range of patient characteristics and evaluation of a broader range of outcomes over a longer period than randomized trials. However, clinicians are often reluctant to incorporate the findings of observational studies into clinical practice. Reason for uncertainty regarding observational studies include a lack of familiarity with observational research methods, occasional disagreements between results of observational studies and randomized trials, the perceived risk of spurious results from systematic bias, and prior teaching that randomized trials are the most reliable source of medical evidence. We propose that a better understanding...

Purpose To quantify the coinciding improvement in the clinical diagnosis of sepsis, its documentation in the electronic health records, and subsequent medical coding of sepsis for billing purposes in recent years. Methods We examined 98,267 hospitalizations in 66,208 patients who met systemic inflammatory response syndrome criteria at a tertiary care center from 2008 to 2012. We used g-computation to estimate the causal effect of the year of hospitalization on receiving an International Classification of Diseases, Ninth Revision, Clinical Modification discharge diagnosis code for sepsis by estimating changes in the probability of getting diagnosed and coded for sepsis during the study period. Results When adjusted for demographics, Charlson-Deyo comorbidity index, blood culture frequency per hospitalization, and intensive care unit admission, the causal risk difference for receiving a discharge code for sepsis per 100 hospitalizations with systemic inflammatory response syndrome, had the hospitalization occurred in 2012, was estimated to be 3.9% (95% confidence interval [CI], 3.8%–4.0%), 3.4% (95% CI, 3.3%–3.5%), 2.2% (95% CI, 2.1%–2.3%), and 0.9% (95% CI, 0.8%–1.1%) from 2008 to 2011, respectively. Conclusions Patients with similar characteristics and risk factors had a higher of probability of getting diagnosed, documented, and coded for sepsis in 2012 than in previous years, which contributed to an apparent increase in sepsis incidence.

The area under the receiver operating characteristic (ROC) curve (AUC) is used as a performance metric for quantitative tests. Although multiple biomarkers may be available for diagnostic or screening purposes, diagnostic accuracy is often assessed individually rather than in combination. In this paper, we consider the interesting problem of combining multiple biomarkers for use in a single diagnostic criterion with the goal of improving the diagnostic accuracy above that of an individual biomarker. The diagnostic criterion created from multiple biomarkers is based on the predictive probability of disease, conditional on given multiple biomarker outcomes. If the computed predictive probability exceeds a specified cutoff, the corresponding subject is allocated as 'diseased'. This defines a standard diagnostic criterion that has its own ROC curve, namely, the combined ROC (cROC). The AUC metric for cROC, namely, the combined AUC (cAUC), is used to compare the predictive criterion based on multiple biomarkers to one based on fewer biomarkers. A multivariate random-effects model is proposed for modeling multiple normally distributed dependent scores. Bayesian methods for estimating ROC curves and corresponding (marginal) AUCs are developed when a perfect reference standard is not available. In addition, cAUCs are computed to compare the accuracy of different combinations of biomarkers for diagnosis. The methods are evaluated using simulations and are applied to data for Johne's disease (paratuberculosis) in cattle.

Purpose To estimate the accuracy of two algorithms to identify cholecystectomy procedures using International Classification of Diseases, 9th Edition, Clinical Modification (ICD-9-CM) and Current Procedural Terminology (CPT-4) codes in administrative data. Methods Private insurer medical claims for 30 853 patients 18–64 years with an inpatient hospitalization between 2006 and 2010, as indicated by providers/facilities place of service in addition to room and board charges, were cross-classified according to the presence of codes for cholecystectomy. The accuracy of ICD-9-CM- and CPT-4-based algorithms was estimated using a Bayesian latent class model. Results The sensitivity and specificity were 0.92 [probability interval (PI): 0.92, 0.92] and 0.99 (PI: 0.97, 0.99) for ICD-9-CM-, and 0.93 (PI: 0.92, 0.93) and 0.99 (PI: 0.97, 0.99) for CPT-4-based algorithms, respectively. The parallel-joint scheme, where positivity of either algorithm was considered a positive outcome, yielded a sensitivity and specificity of 0.99 (PI: 0.99, 0.99) and 0.97 (PI: 0.95, 0.99), respectively. Conclusions Both ICD-9-CM- and CPT-4-based algorithms had high sensitivity to identify cholecystectomy procedures in administrative data when used individually and especially in a parallel-joint approach. Copyright © 2015 John Wiley & Sons, Ltd.

Re-institution of reflex testing of stool submitted for Clostridium difficile testing for vancomycin-resistant enterococci (VRE) reduced the incidence of healthcare-associated VRE bacteremia and bacteriuria compared to when testing was not in place (1.9 versus 3.3 cases per 10,000 patient days when testing was not in use).

Purpose Past studies of sepsis epidemiology did not address misclassification bias due to imperfect verification of sepsis detection methods to estimate the true prevalence. Methods We examined 273,126 hospitalizations from 2008 to 2012 at a tertiary-care center to develop surveillance-aimed sepsis detection criteria, based on the presence of the sepsis-explicit International Classification of Diseases, Ninth Revision, Clinical Modification codes (995.92 or 785.52), blood culture orders, and antibiotics administration. We used Bayesian multinomial latent class models to estimate the true prevalence of sepsis, while adjusting for the imperfect sensitivity and specificity and the conditional dependence among the individual criteria. Results The apparent annual prevalence of sepsis hospitalizations based on explicit International Classification of Diseases, Ninth Revision, Clinical Modification codes were 1.5%, 1.4%, 1.6%, 2.2%, and 2.5% for the years 2008 to 2012. Bayesian posterior estimates for the true prevalence of sepsis suggested that it remained stable from 2008, 19.2% (95% credible interval [CI]: 17.9%, 22.9%), to 2012, 17.8% (95% CI: 16.8%, 20.2%). The sensitivity of sepsis-explicit codes, however, increased from 7.6% (95% CI: 6.4%, 8.4%) in 2008 to 13.8% (95% CI: 12.2%, 14.9%) in 2012. Conclusions The true prevalence of sepsis remained high, but stable despite an increase in the sensitivity of sepsis-explicit codes in administrative data.

BACKGROUND The long-term and cumulative effect of multiple episodes of bacteremia and sepsis across multiple hospitalizations on the development of cardiovascular (CV) events is uncertain. METHODS We conducted a longitudinal study of 156 380 hospitalizations in 47 009 patients (≥18 years old) who had at least 2 inpatient admissions at an academic tertiary care center in St Louis, Missouri, from 1 January 2008 through 31 December 2012. We used marginal structural models, estimated by inverse probability weighting (IPW) of bacteremia or sepsis and IPW of censoring, to estimate the marginal causal effects of bacteremia and sepsis on developing the first observed incident CV event, including stroke, transient ischemic attack, and myocardial infarction (MI), during the study period. RESULTS Bacteremia and sepsis occurred during 4923 (3.1%) and 5544 (3.5%) hospitalizations among 3932 (8.4%) and 4474 (9.5%) patients, respectively. CV events occurred in 414 (10.5%) and 538 (12.0%) patients with prior episodes of bacteremia or sepsis, respectively, vs 3087 (7.2%) and 2963 (7.0%) patients without prior episodes of bacteremia or sepsis. The causal odds of experiencing a CV event was 1.52-fold (95% confidence interval [CI], 1.21- to 1.90-fold) and 2.39-fold (95% CI, 1.88- to 3.03-fold) higher in patients with prior instances of bacteremia or sepsis, respectively, compared to those without. Prior instances of septic shock resulted in a 6.91-fold (95% CI, 5.34- to 8.93-fold) increase in the odds of MI. CONCLUSIONS Prior instances of bacteremia and sepsis substantially increase the 5-year risk of CV events.

BACKGROUND Recent reports using administrative claims data suggest the incidence of community- and hospital-onset sepsis is increasing. Whether this reflects changing epidemiology, more effective diagnostic methods, or changes in physician documentation and medical coding practices is unclear. METHODS We performed a temporal-trend study from 2008 to 2012 using administrative claims data and patient-level clinical data of adult patients admitted to Barnes-Jewish Hospital in St. Louis, Missouri. Temporal-trend and annual percent change were estimated using regression models with autoregressive integrated moving average errors. RESULTS We analyzed 62,261 inpatient admissions during the 5-year study period. 'Any SIRS' (i.e., SIRS on a single calendar day during the hospitalization) and 'multi-day SIRS' (i.e., SIRS on 3 or more calendar days), which both use patient-level data, and medical coding for sepsis (i.e., ICD-9-CM discharge diagnosis codes 995.91, 995.92, or 785.52) were present in 35.3 %, 17.3 %, and 3.3 % of admissions, respectively. The incidence of admissions coded for sepsis increased 9.7 % (95 % CI: 6.1, 13.4) per year, while the patient data-defined events of 'any SIRS' decreased by 1.8 % (95 % CI: -3.2, -0.5) and 'multi-day SIRS' did not change significantly over the study period. Clinically-defined sepsis (defined as SIRS plus bacteremia) and severe sepsis (defined as SIRS plus hypotension and bacteremia) decreased at statistically significant rates of 5.7 % (95 % CI: -9.0, -2.4) and 8.6 % (95 % CI: -4.4, -12.6) annually. All-cause mortality, SIRS mortality, and SIRS and clinically-defined sepsis case fatality did not change significantly during the study period. Sepsis mortality, based on ICD-9-CM codes, however, increased by 8.8 % (95 % CI: 1.9, 16.2) annually. CONCLUSIONS The incidence of sepsis, defined by ICD-9-CM codes, and sepsis mortality increased steadily without a concomitant increase in SIRS or clinically-defined sepsis. Our results highlight the need to develop strategies to integrate clinical patient-level data with administrative data to draw more accurate conclusions about the epidemiology of sepsis.

To estimate the accuracy of two algorithms to identify cholecystectomy procedures using International Classification of Diseases, 9th Edition, Clinical Modification (ICD-9-CM) and Current Procedural Terminology (CPT-4) codes in administrative data.Private insurer medical claims for 30 853 patients 18-64 years with an inpatient hospitalization between 2006 and 2010, as indicated by providers/facilities place of service in addition to room and board charges, were cross-classified according to the presence of codes for cholecystectomy. The accuracy of ICD-9-CM- and CPT-4-based algorithms was estimated using a Bayesian latent class model.The sensitivity and specificity were 0.92 [probability interval (PI): 0.92, 0.92] and 0.99 (PI: 0.97, 0.99) for ICD-9-CM-, and 0.93 (PI: 0.92, 0.93) and 0.99 (PI: 0.97, 0.99) for CPT-4-based algorithms, respectively. The parallel-joint scheme, where positivity of either algorithm was considered a positive outcome, yielded a sensitivity and specificity of 0.99 (PI: 0.99, 0.99) and 0.97 (PI: 0.95, 0.99), respectively.Both ICD-9-CM- and CPT-4-based algorithms had high sensitivity to identify cholecystectomy procedures in administrative data when used individually and especially in a parallel-joint approach.

To identify events that could predict province-level frequency of foot-and-mouth disease (FMD) outbreaks in Iran, 5707 outbreaks reported from April 1995 to March 2002 were studied. A zero-inflated negative binomial model was used to estimate the probability of a ‘no-outbreak’ status and the number of outbreaks in a province, using the number of previous occurrences of FMD for the same or adjacent provinces and season as covariates. For each province, the probability of observing no outbreak was negatively associated with the number of outbreaks in the same province in the previous month (odds ratio [OR] = 0.06, 95% confidence interval [CI]: 0.01, 0.30) and in ‘the second previous month’ (OR = 0.10, 95% CI: 0.02, 0.51), the total number of outbreaks in the second previous month in adjacent provinces (OR = 0.57, 95% CI: 0.36, 0.91) and the season (winter [OR = 0.18, 95% CI: 0.06, 0.55] and spring [OR = 0.27, 95% CI: 0.09, 0.81], compared with summer). The expected number of outbreaks in a province was positively associated with number of outbreaks in the same province in previous month (coefficient [coef] = 0.74, 95% CI: 0.66, 0.82) and in the second previous month (coef = 0.23, 95% CI: 0.16, 0.31), total number of outbreaks in adjacent provinces in the previous month (coef = 0.32, 95% CI: 0.22, 0.41) and season (fall [coef = 0.20, 95% CI: 0.07, 0.33] and spring [coef = 0.18, 95% CI: 0.05, 0.31], compared to summer); however, number of outbreaks was negatively associated with the total number of outbreaks in adjacent provinces in the second previous month (coef = −0.19, 95% CI: −0.28, −0.09). The findings indicate that the probability of an outbreak (and the expected number of outbreaks if any) may be predicted based on previous province information, which could help decision-makers allocate resources more efficiently for province-level disease control measures. Further, the study illustrates use of zero inflated negative binomial model to study diseases occurrence where disease is infrequently observed.

The area under the receiver operating characteristic (ROC) curve (AUC) is used as a performance metric for quantitative tests. Although multiple biomarkers may be available for diagnostic or screening purposes, diagnostic accuracy is often assessed individually rather than in combination. In this paper, we consider the interesting problem of combining multiple biomarkers for use in a single diagnostic criterion with the goal of improving the diagnostic accuracy above that of an individual biomarker. The diagnostic criterion created from multiple biomarkers is based on the predictive probability of disease, conditional on given multiple biomarker outcomes. If the computed predictive probability exceeds a specified cutoff, the corresponding subject is allocated as 'diseased'. This defines a standard diagnostic criterion that has its own ROC curve, namely, the combined ROC (cROC). The AUC metric for cROC, namely, the combined AUC (cAUC), is used to compare the predictive criterion based on multiple biomarkers to one based on fewer biomarkers. A multivariate random-effects model is proposed for modeling multiple normally distributed dependent scores. Bayesian methods for estimating ROC curves and corresponding (marginal) AUCs are developed when a perfect reference standard is not available. In addition, cAUCs are computed to compare the accuracy of different combinations of biomarkers for diagnosis. The methods are evaluated using simulations and are applied to data for Johne's disease (paratuberculosis) in cattle.

s / Annals of Epidemiology 24 (2014) 682e702 691 cross-classified mixed-effects model was fit, treating drugs as nested within drug-classes, and patients and drugs as cross-classified. Results: Adherence varied within, but overlapped between drug-classes; for example, average adherence to statins ranged from 64.8% (rosuvastatin) to 75.1% (pravastatin), and beta-blockers from 63.8% (timolol) to 79.6% (nadolol). Most variance in adherence was attributable to time-variant drugand patient-characteristics (65.9%), followed by time-invariant drug-specific attributes (29.2%); invariant patient characteristics and drug-class together explained 4.9%. Substantial variance within drug-class remained after adjusting for age, years living with diabetes, illness severity, number of medications, healthcare access, cost, socioeconomic status, sex, routes of administration, adverse events reported by drug, and combination-drugs. Conclusions: The majority of the variation in adherence is attributable to modifiable factors, including the actual drug prescribed. Adjusted differences in adherence by drug within classes may be used to select drugs with better adherence profiles. Further research is needed to understand variation in adherence at the drug-level.