Fredrik A. Dahl

Obesity and osteoarthritis in knee, hip and/or hand: An epidemiological study in the general population with 10 years follow-up

BackgroundObesity is one of the most important risk factors for osteoarthritis (OA) in knee(s). However, the relationship between obesity and OA in hand(s) and hip(s) remains controversial and needs further investigation. The purpose of this study was to investigate the impact of obesity on incident osteoarthritis (OA) in hip, knee, and hand in a general population followed in 10 years.MethodsA total of 1854 people aged 24–76 years in 1994 participated in a Norwegian study on musculoskeletal pain in both 1994 and 2004. Participants with OA or rheumatoid arthritis in 1994 and those above 74 years in 1994 were excluded, leaving n = 1675 for the analyses. The main outcome measure was OA diagnosis at follow-up based on self-report. Obesity was defined by a body mass index (BMI) of 30 and above.ResultsAt 10-years follow-up the incidence rates were 5.8% (CI 4.3–7.3) for hip OA, 7.3% (CI 5.7–9.0) for knee OA, and 5.6% (CI 4.2–7.1) for hand OA. When adjusting for age, gender, work status and leisure time activities, a high BMI (> 30) was significantly associated with knee OA (OR 2.81; 95%CI 1.32–5.96), and a dose-response relationship was found for this association. Obesity was also significantly associated with hand OA (OR 2.59; 1.08–6.19), but not with hip OA (OR 1.11; 0.41–2.97). There was no statistically significant interaction effect between BMI and gender, age or any of the other confounding variables.ConclusionA high BMI was significantly associated with knee OA and hand OA, but not with hip OA.

Laparoscopic Lavage vs Primary Resection for Acute Perforated Diverticulitis: The SCANDIV Randomized Clinical Trial

IMPORTANCE Perforated colonic diverticulitis usually requires surgical resection, which is associated with significant morbidity. Cohort studies have suggested that laparoscopic lavage may treat perforated diverticulitis with less morbidity than resection procedures. OBJECTIVE To compare the outcomes from laparoscopic lavage with those for colon resection for perforated diverticulitis. DESIGN, SETTING, AND PARTICIPANTS Multicenter, randomized clinical superiority trial recruiting participants from 21 centers in Sweden and Norway from February 2010 to June 2014. The last patient follow-up was in December 2014 and final review and verification of the medical records was assessed in March 2015. Patients with suspected perforated diverticulitis, a clinical indication for emergency surgery, and free air on an abdominal computed tomography scan were eligible. Of 509 patients screened, 415 were eligible and 199 were enrolled. INTERVENTIONS Patients were assigned to undergo laparoscopic peritoneal lavage (n = 101) or colon resection (n = 98) based on a computer-generated, center-stratified block randomization. All patients with fecal peritonitis (15 patients in the laparoscopic peritoneal lavage group vs 13 in the colon resection group) underwent colon resection. Patients with a pathology requiring treatment beyond that necessary for perforated diverticulitis (12 in the laparoscopic lavage group vs 13 in the colon resection group) were also excluded from the protocol operations and treated as required for the pathology encountered. MAIN OUTCOMES AND MEASURES The primary outcome was severe postoperative complications (Clavien-Dindo score >IIIa) within 90 days. Secondary outcomes included other postoperative complications, reoperations, length of operating time, length of postoperative hospital stay, and quality of life. RESULTS The primary outcome was observed in 31 of 101 patients (30.7%) in the laparoscopic lavage group and 25 of 96 patients (26.0%) in the colon resection group (difference, 4.7% [95% CI, -7.9% to 17.0%]; P = .53). Mortality at 90 days did not significantly differ between the laparoscopic lavage group (14 patients [13.9%]) and the colon resection group (11 patients [11.5%]; difference, 2.4% [95% CI, -7.2% to 11.9%]; P = .67). The reoperation rate was significantly higher in the laparoscopic lavage group (15 of 74 patients [20.3%]) than in the colon resection group (4 of 70 patients [5.7%]; difference, 14.6% [95% CI, 3.5% to 25.6%]; P = .01) for patients who did not have fecal peritonitis. The length of operating time was significantly shorter in the laparoscopic lavage group; whereas, length of postoperative hospital stay and quality of life did not differ significantly between groups. Four sigmoid carcinomas were missed with laparoscopic lavage. CONCLUSIONS AND RELEVANCE Among patients with likely perforated diverticulitis and undergoing emergency surgery, the use of laparoscopic lavage vs primary resection did not reduce severe postoperative complications and led to worse outcomes in secondary end points. These findings do not support laparoscopic lavage for treatment of perforated diverticulitis. TRIAL REGISTRATION clinicaltrials.gov Identifier: NCT01047462.

Effectiveness of a short course in clinical communication skills for hospital doctors: results of a crossover randomized controlled trial (ISRCTN22153332).

OBJECTIVE To test the hypothesis that a 20-h communication skills course based on the Four Habits model can improve doctor-patient communication among hospital employed doctors across specialties. METHODS Crossover randomized controlled trial in a 500-bed hospital with interventions at different time points in the two arms. Assessments were video-based and blinded. Intervention consisted of 20 h of communication training, containing alternating plenary with theory/debriefs and practical group sessions with role-plays tailored to each doctor. RESULTS Of 103 doctors asked to participate, 72 were included, 62 received the intervention, 51 were included in the main analysis, and another six were included in the intention-to-treat analysis. We found an increase in the Four Habits Coding Scheme of 7.5 points (p = 0.01, 95% confidence interval 1.6-13.3), fairly evenly distributed on subgroups. Baseline score (SD) was 60.3 (9.9). Global patient satisfaction did not change, neither did average encounter duration. CONCLUSION Utilizing an outpatient-clinic training model developed in the US, we demonstrated that a 20-h course could be generalized across medical and national cultures, indicating improvement of communication skills among hospital doctors. PRACTICE IMPLICATIONS The Four Habits model is suitable for communication-training courses in hospital settings. Doctors across specialties can attend the same course.

Post-Processing Posterior Predictive p Values

This article addresses issues of model criticism and model comparison in Bayesian contexts, and focuses on the use of the so-called posterior predictive p values (ppp). These involve a general discrepancy or conflict measure and depend on the prior, the model, and the data. They are used in statistical practice to quantify the degree of surprise or conflict in data and to compare different combinations of prior and model. The distribution of such ppp values is far from uniform however, as we demonstrate for different models, making their interpretation and comparison a difficult matter. We propose a natural calibration of the ppp values, where the resulting cppp values are uniform on the unit interval under model conditions. The cppp values, which in general rely on a double-simulation scheme for their computation, may then be used to assess and compare different priors and models. Our methods also make it possible to compare parametric and nonparametric model specifications, in that genuine “measures of surprise” are put on the same canonical uniform scale. We illustrate our techniques for some applications to real data. We also present supplementing theoretical results on various properties of the ppp and cppp.

Data splitting as a countermeasure against hypothesis fishing: with a case study of predictors for low back pain

There is growing concern in the scientific community that many published scientific findings may represent spurious patterns that are not reproducible in independent data sets. A reason for this is that significance levels or confidence intervals are often applied to secondary variables or sub-samples within the trial, in addition to the primary hypotheses (multiple hypotheses). This problem is likely to be extensive for population-based surveys, in which epidemiological hypotheses are derived after seeing the data set (hypothesis fishing). We recommend a data-splitting procedure to counteract this methodological problem, in which one part of the data set is used for identifying hypotheses, and the other is used for hypothesis testing. The procedure is similar to two-stage analysis of microarray data. We illustrate the process using a real data set related to predictors of low back pain at 14-year follow-up in a population initially free of low back pain. “Widespreadness” of pain (pain reported in several other places than the low back) was a statistically significant predictor, while smoking was not, despite its strong association with low back pain in the first half of the data set. We argue that the application of data splitting, in which an independent party handles the data set, will achieve for epidemiological surveys what pre-registration has done for clinical studies.

A randomised comparison of a four- and a five-point scale version of the Norwegian Function Assessment Scale

BackgroundThere is variation in the number of response alternatives used within health-related questionnaires. This study compared a four-and a five-point scale version of the Norwegian Function Assessment Scale (NFAS) by evaluating data quality, internal consistency and validity.MethodsAll inhabitants in seven birth cohorts in the Ullensaker municipality of Norway were approached by means of a postal questionnaire. The NFAS was included as part of The Ullensaker Study 2004. The instrument comprises 39 items derived from the activities/participation component in the International Classification for Functioning, Disabilities and Health (ICF). The sample was computer-randomised to either the four-point or the five-point scale version.ResultsBoth versions of the NFAS had acceptable response rates and good data quality and internal consistency. The five-point scale version had better data quality in terms of missing data, end effects at the item and scale level, as well as higher levels of internal consistency. Construct validity was acceptable for both versions, demonstrated by correlations with instruments assessing similar aspects of health and comparisons with groups of individuals known to differ in their functioning according to existing evidence.ConclusionData quality, internal consistency and discriminative validity suggest that the five-point scale version should be used in future applications.

Factors associated with a prolonged length of stay after acute exacerbation of chronic obstructive pulmonary disease (AECOPD).

Background Early identification of patients with a prolonged stay due to acute exacerbation of chronic obstructive pulmonary disease (COPD) may reduce risk of adverse event and treatment costs. This study aimed to identify predictors of prolonged stay after acute exacerbation of COPD based on variables on admission; the study also looked to establish a prediction model for length of stay (LOS). Methods We extracted demographic and clinical data from the medical records of 599 patients discharged after an acute exacerbation of COPD between March 2006 and December 2008 at Oslo University Hospital, Aker. We used logistic regression analyses to assess predictors of a length of stay above the 75th percentile and assessed the area under the receiving operating characteristic curve to evaluate the model’s performance. Results We included 590 patients (54% women) aged 73.2±10.8 years (mean ± standard deviation) in the analyses. Median LOS was 6.0 days (interquartile range [IQR] 3.5–11.0). In multivariate analysis, admission between Thursday and Saturday (odds ratio [OR] 2.24 [95% CI 1.60–3.51], P<0.001), heart failure (OR 2.26, 95% CI 1.34–3.80), diabetes (OR 1.90, 95% CI 1.07–3.37), stroke (OR 1.83, 95% CI 1.04–3.21), high arterial PCO2 (OR 1.26 [95% CI 1.13–1.41], P<0.001), and low serum albumin level (OR 0.92 [95% CI 0.87–0.97], P=0.001) were associated with a LOS >11 days. The statistical model had an area under the receiver operating characteristic curve of 0.73. Conclusion Admission between Thursday and Saturday, heart failure, diabetes, stroke, high arterial PCO2, and low serum albumin level were associated with a prolonged LOS. These findings may help physicians to identify patients that will need a prolonged LOS in the early stages of admission. However, the predictive model exhibited suboptimal performance and hence is not ready for clinical use.

Interrater reliability for the Four Habits Coding Scheme as part of a randomized controlled trial

OBJECTIVE To describe the process for developing interrater reliability (IRR) for the Four Habits Coding Scheme (4HCS) for a heterogeneous material as part of a randomized controlled trial. METHODS Videotapes from 497 hospital encounters involving 71 doctors from most clinical specialties were collected. Four experienced psychology students were trained as raters. We calculated Pearsons r and the intraclass correlation (ICC) on the total score across consecutive samples of twenty videos, and Pearsons r on single videos across items in the initial coding phase. RESULTS After 18h of training and one rating session, the total score Pearsons r and ICC exceeded .70 for all pairs of raters. Across items within single videos, the Pearsons r was never below 0.60 after the first 50 videos. At item and habit level Pearsons r remained unsatisfactory for some rater pairs mostly due to low variance on some items. CONCLUSION Based on the evaluation of the effect of communication skills training via a total score, IRR was satisfactory for the 4HCS as applied to heterogeneous material. However, good reliability at item level was difficult to achieve. PRACTICE IMPLICATIONS 4HCS may be used as an outcome measure for clinical communication skills in randomized controlled trials.

Simulating the effect of physician triage in the emergency department of Akershus University Hospital

The Norwegian Board of Health Supervision has strongly recommended that all hospitals need to take action to improve the long waiting times before patients are seen by a physician in the Hospital Emergency Department. Akershus University Hospital has complied with this by introducing physician triage every weekday from 10am to 7pm. Because it is difficult to see the influence this has had on the patient flow in the ED, the Hospital Research Department has developed two simulation models to estimate the effect on patient waiting time by replacing nurse triage with that of a physician. The results of the simulations show that the waiting time for an initial physician evaluation was reduced from 117 minutes to 26 minutes, while the waiting time for a physician examination was reduced only by 7 minutes. The total waiting time in the ED was reduced from 297 to 288 minutes when introducing physician triage.

The Lagging Anchor Algorithm: Reinforcement Learning in Two-Player Zero-Sum Games with Imperfect Information

The article describes a gradient search based reinforcement learning algorithm for two-player zero-sum games with imperfect information. Simple gradient search may result in oscillation around solution points, a problem similar to the “Crawford puzzle”. To dampen oscillations, the algorithm uses lagging anchors, drawing the strategy state of the players toward a weighted average of earlier strategy states. The algorithm is applicable to games represented in extensive form. We develop methods for sampling the parameter gradient of a players performance against an opponent, using temporal-difference learning. The algorithm is used successfully for a simplified poker game with infinite sets of pure strategies, and for the air combat game Campaign, using neural nets. We prove exponential convergence of the algorithm for a subset of matrix games.