Sample size calculations for skewed distributions

Performance of administrative database frailty instruments in predicting clinical outcomes and cost for patients undergoing transcatheter aortic valve implantation: a historical cohort study

PURPOSE

Frailty instruments may improve prognostic estimates for patients undergoing transcatheter aortic valve implantation (TAVI). Few studies have evaluated and compared the performance of administrative database frailty instruments for patients undergoing TAVI. This study aimed to examine the performance of administrative database frailty instruments in predicting clinical outcomes and costs in patients who underwent TAVI.

METHODS

We conducted a historical cohort study of 3,848 patients aged 66 yr or older who underwent a TAVI procedure in Ontario, Canada from 1 April 2012 to 31 March 2018. We used the Johns Hopkins Adjusted Clinical Group (ACG) frailty indicator and the Hospital Frailty Risk Score (HFRS) to assign frailty status. Outcomes of interest were in-hospital mortality, one-year mortality, rehospitalization, and healthcare costs. We compared the performance of the two frailty instruments with that of a reference model that adjusted baseline covariates and procedural characteristics. Accuracy measures included c-statistics, Akaike information criterion (AIC), Bayesian information criterion (BIC), integrated discrimination improvement (IDI), net reclassification index (NRI), bias, and accuracy of cost estimates.

RESULTS

A total of 863 patients (22.4%) were identified as frail using the Johns Hopkins ACG frailty indicator and 865 (22.5%) were identified as frail using the HFRS. Although agreement between the frailty instruments was fair (Kappa statistic = 0.322), each instrument classified different subgroups as frail. Both the Johns Hopkins ACG frailty indicator (rate ratio [RR], 1.13; 95% confidence interval [CI], 1.06 to 1.20) and the HFRS (RR, 1.14; 95% CI, 1.07 to 1.21) were significantly associated with increased one-year costs. Compared with the reference model, both the Johns Hopkins ACG frailty indicator and HFRS significantly improved NRI for one-year mortality (Johns Hopkins ACG frailty indicator: NRI, 0.160; P < 0.001; HFRS: NRI, 0.146; P = 0.001) and rehospitalization (Johns Hopkins ACG frailty indicator: NRI, 0.201; P < 0.001; HFRS: NRI, 0.141; P = 0.001). These improvements in NRI largely resulted from classification improvement among those who did not experience the event. With one-year mortality, there was a significant improvement in IDI (IDI, 0.003; P < 0.001) with the Johns Hopkins ACG frailty indicator. This improvement in performance resulted from an increase in the mean probability of the event among those with the event.

CONCLUSION

Preoperative frailty assessment may add some predictive value for TAVI outcomes. Use of administrative database frailty instruments may provide small but significant improvements in case-mix adjustment when profiling hospitals for certain outcomes.

Cognition and motor phenotypes in ALS: a retrospective study

Background

Amyotrophic lateral sclerosis (ALS) is phenotypically heterogeneous in motor manifestations, and the extent of upper vs. lower motor neuron involvement is a widespread descriptor. This study aimed to examine cognition across different ALS motor phenotypes.

Methods

ALS patients (N = 124) were classified as classical (N = 66), bulbar (N = 13), predominant-upper motor neuron (PUMN; N = 19), and predominant-lower motor neuron (PLMN; N = 26) phenotypes. Cognition was assessed with the Edinburgh Cognitive and Behavioural ALS Screen (ECAS) and function with the ALS Functional Rating Scale—Revised (ALSFRS-R). Revised ALS-FTD consensus criteria were applied for cognitive/behavioral phenotyping.

Results

Defective ECAS-total scores were detected in all groups — bulbar: 15.4%, classical: 30.3%, PLMN: 23.1%, and PUMN: 36.8%. Classical and PUMN ALS patients performed worse than PLMN ones on ECAS-total, ALS-specific, Fluency, and Executive measures. No other difference was detected. Worse ASLFRS-R scores correlated with poorer ECAS-total scores in classical ALS patients.

Conclusions

Frontotemporal cognitive deficits are more prevalent in PUMN and classical ALS and linked to disease severity in the latter, but occur also in PLMN phenotypes.

Supplementary information

The online version contains supplementary material available at 10.1007/s10072-022-06157-x

Quantifying the sources of variation in eosinophilia among Scottish blackface lambs with mixed, predominantly Teladorsagia circumcincta nematode infection

Eosinophils play a key role in defence against gastrointestinal nematodes. There is considerable variation among animals in the intensity of eosinophilia following nematode infection. However, the statistical distribution of eosinophils among animals has still to be determined. A better description of the variation among animals could provide biological insight and determine the most appropriate way to analyse the effect of eosinophils. We estimated blood eosinophil numbers in a flock of Scottish Blackface sheep that were naturally exposed to mixed, predominantly Teladorsagia circumcincta infection. Three of the four eosinophil counts were better described by a gamma distribution than by a lognormal distribution. The scale and shape parameters of the gamma distribution varied over time. Eosinophil counts differed among animals kept on separate fields before weaning and between singletons and twins but were not significantly different between years and genders. Eosinophil counts also differed among offspring from different sires and dams. The parameters of the gamma distribution were used to enable a power analysis. Large numbers of animals were required to reliably detect even large differences between two groups. These results indicate that methods appropriate for gamma distributions, such as generalized linear mixed models, will provide more reliable inferences than traditional methods of analysis and experimental design.

Overgeneralization of fear, but not avoidance, following acute stress

Research has demonstrated the spreading of fear from threat-related stimuli to perceptually similar, but innocuous, stimuli. Less is known, however, about the generalization of avoidance behavior. Given that stress is known to affect learning and memory, we were interested in the effect of acute stress on (over)generalization of fear and avoidance responses. On the first day, one geometrical shape was paired with a mild electrical stimulus (CS+), whereas another shape was not (CS-). One day later, after participants had been exposed to the Maastricht Acute Stress Test or a control task, generalization of avoidance responses and fear (shock expectancy and skin conductance responses) was tested to a range of perceptual generalization stimuli. Generalization gradients were observed across different outcome measures. Stress enhanced generalization of shock expectancy to the stimulus most similar to the CS+. Our findings confirm that stress can affect the generalization of fear, but further studies are warranted.

A Priori Sample Size Determination for the Number of Subjects in an EEG Experiment

This paper represents a first attempt to perform a priori sample size determination from a "historic" Electroencephalography (EEG) dataset. The importance of adequate sample size is firstly highlighted, and evidence is given against the use of normal distribution for such computations, when the data cannot be assumed to be Gaussian. The "historic" dataset is then thoroughly examined to determine the least less likely underlying distribution for the desired phenomenon, in this case the spontaneous blinks potential. Two Monte Carlo simulations, using different distribution assumptions, are subsequently computed to estimate the a priori minimum sample size. Finally, these choices are discussed considering practical limitations, as well as the computational differences for other phenomena to study.

An integrative care bundle to prevent surgical site infections among surgical hip patients: A retrospective cohort study

BACKGROUND

Surgical site infections (SSIs) following hip fracture surgeries have profound clinical and economic implications. The study aims to analyze the effect of an integrative SSI prevention care bundle on the SSI incidence among surgical hip patients. The study also aims to examine the association between SSI and mortality, readmission, duration of hospitalization, and hospital cost.

METHODS

A retrospective cohort study was carried out to assess the incidence of SSI in an acute hip unit in Singapore from January 2015 to September 2017. Patients who developed SSI fell into the exposure group, whereas patients without SSI fell into the nonexposure group. A comparison of the incidence of mortality, readmission, length of stay, and inpatient bill size was conducted between the 2 groups.

RESULTS

Among 758 hip surgeries performed during the study period, 14 (1.8%) SSIs were documented. Compared with patients with no SSI, patients with SSI were 4.27times more likely to be readmitted within 30days, had 2.47times longer length of stay, and 2.15times the inpatient bill size.

CONCLUSIONS

An integrative care bundle that capitalizes on the expertise of a multidisciplinary team has promoted shared responsibility and proven to be effective in preventing SSIs while contributing to better patient outcomes within the unit.

Proactive expert system intervention to prevent or quit at-risk alcohol use (PRINT): study protocol of a randomized controlled trial

BACKGROUND

The population impact of alcohol screening and brief intervention might be increased by approaching an entire population rather than individuals at high risk only. The aim is to present the protocol of the study "Testing a proactive expert system intervention to prevent and to quit at-risk alcohol use" (PRINT) which tests the efficacy of a computer-based brief intervention (i) to elicit drinking reductions among persons with at-risk alcohol use and (ii) to prevent at-risk alcohol use among current low-risk drinkers.

METHODS/DESIGN

The PRINT study is a two-arm randomized controlled trial with a 12-month follow-up. A total of 1648 participants will be proactively recruited in the waiting area of a municipal registry office. All 18- to 64-year-old persons with past year alcohol use will be randomized to either the intervention group or the control group. Participants in the intervention group will receive computer-generated individualized feedback letters at baseline, month 3, and month 6. Participants in the control group will receive assessment only. The primary outcome is the change in the number of drinks per day from baseline to month 12.

DISCUSSION

We expect to provide a computer-based brief alcohol intervention that is appropriate for a wide range of people with alcohol use regardless of their initial alcohol-risk level. The intervention might have the potential to decrease alcohol use and alcohol-related problems on a population level at low costs.

TRIAL REGISTRATION

German Clinical Trials Register: DRKS00014274 (date of registration: 2018/03/12).

Neonatal resuscitation using a laryngeal mask airway: a randomised trial in Uganda

OBJECTIVE

Mortality rates from birth asphyxia in low-income countries remain high. Face mask ventilation (FMV) performed by midwives is the usual method of resuscitating neonates in such settings but may not always be effective. The i-gel is a cuffless laryngeal mask airway (LMA) that could enhance neonatal resuscitation performance. We aimed to compare LMA and face mask (FM) during neonatal resuscitation in a low-resource setting.

SETTING

Mulago National Referral Hospital, Kampala, Uganda.

DESIGN

This prospective randomised clinical trial was conducted at the labour ward operating theatre. After a brief training on LMA and FM use, infants with a birth weight >2000 g and requiring positive pressure ventilation at birth were randomised to resuscitation by LMA or FM. Resuscitations were video recorded.

MAIN OUTCOME MEASURES

Time to spontaneous breathing.

RESULTS

Forty-nine (24 in the LMA and 25 in the FM arm) out of 50 enrolled patients were analysed. Baseline characteristics were comparable between the two arms. Time to spontaneous breathing was shorter in LMA arm than in FM arm (mean 153 s (SD±59) vs 216 s (SD±92)). All resuscitations were effective in LMA arm, whereas 11 patients receiving FM were converted to LMA because response to FMV was unsatisfactory. There were no adverse effects.

CONCLUSION

A cuffless LMA was more effective than FM in reducing time to spontaneous breathing. LMA seems to be safe and effective in clinical practice after a short training programme. Its potential benefits on long-term outcomes need to be assessed in a larger trial.

CLINICAL TRIAL REGISTRY

This trial was registered in https://clinicaltrials.gov, with registration number NCT02042118.

Home versus hospital immunoglobulin treatment for autoimmune neuropathies: A cost minimization analysis

Background

Prior clinical trials have suggested that home-based Ig treatment in multifocal motor neuropathy (MMN) and chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) and its variant Lewis-Sumner syndrome (LSS) is safe and effective and is less costly than hospital-administered intravenous immunoglobulin (IVIg).

Methods

A French prospective, dual-center, cost minimization analysis was carried out to evaluate IVIg administration (5% concentrated) at home versus in hospital with regard to costs, patients' autonomy, and patients' quality of life. The primary endpoint was the overall cost of treatment, and we adopted the perspective of the payer (French Social Health Insurance).

Results

Twenty-four patients aged 52.3 (12.2) years were analyzed: nine patients with MMN, eight with CIDP, and seven with LSS. IVIg (g/kg) dosage was 1.51 ± 0.43 in hospital and 1.52 ± 0.4 at home. Nine-month total costs per patient extrapolated to 1 year of treatment were €48,189 ± 26,105 versus €91,798 ± 51,125 in the home and hospital groups, respectively (p < .0001). The most frequently reported factors for choosing home treatment were the good tolerance and absence of side effects of IVIg administration, as well as a good understanding of the advantages and drawbacks of home treatment (75% of respondents). The mRankin scores before and after switch to home treatment were 1.61 ± 0.72 and 1.36 ± 0.76, respectively (p = .027).

Discussion

The switch from hospital-based to home-based IVIg treatment for patients with immune neuropathy represents potentially significant savings in the management of the disease.

Sample size calculations based on a difference in medians for positively skewed outcomes in health care studies

Background

In healthcare research, outcomes with skewed probability distributions are common. Sample size calculations for such outcomes are typically based on estimates on a transformed scale (e.g. log) which may sometimes be difficult to obtain. In contrast, estimates of median and variance on the untransformed scale are generally easier to pre-specify. The aim of this paper is to describe how to calculate a sample size for a two group comparison of interest based on median and untransformed variance estimates for log-normal outcome data.

Methods

A log-normal distribution for outcome data is assumed and a sample size calculation approach for a two-sample t-test that compares log-transformed outcome data is demonstrated where the change of interest is specified as difference in median values on the untransformed scale. A simulation study is used to compare the method with a non-parametric alternative (Mann-Whitney U test) in a variety of scenarios and the method is applied to a real example in neurosurgery.

Results

The method attained a nominal power value in simulation studies and was favourable in comparison to a Mann-Whitney U test and a two-sample t-test of untransformed outcomes. In addition, the method can be adjusted and used in some situations where the outcome distribution is not strictly log-normal.

Conclusions

We recommend the use of this sample size calculation approach for outcome data that are expected to be positively skewed and where a two group comparison on a log-transformed scale is planned. An advantage of this method over usual calculations based on estimates on the log-transformed scale is that it allows clinical efficacy to be specified as a difference in medians and requires a variance estimate on the untransformed scale. Such estimates are often easier to obtain and more interpretable than those for log-transformed outcomes.

Electronic supplementary material

The online version of this article (doi:10.1186/s12874-017-0426-1) contains supplementary material, which is available to authorized users.

Protocol for assessing maternal, environmental and epigenetic risk factors for dental caries in children

Background

Expenditure on dental and oral health services in Australia is $3.4 billion AUD annually. This is the sixth highest health cost and accounts for 7 % of total national health expenditure. Approximately 49 % of Australian children aged 6 years have caries experience in their deciduous teeth and this is rising. The aetiology of dental caries involves a complex interplay of individual, behavioural, social, economic, political and environmental conditions, and there is increasing interest in genetic predisposition and epigenetic modification.

Methods

The Oral Health Sub-study; a cross sectional study of a birth cohort began in November 2012 by examining mothers and their children who were six years old by the time of initiation of the study, which is ongoing. Data from detailed questionnaires of families from birth onwards and data on mothers’ knowledge, attitudes and practices towards oral health collected at the time of clinical examination are used. Subjects’ height, weight and mid-waist circumference are taken and Body Mass Index (BMI) computed, using an electronic Bio-Impedance balance. Dental caries experience is scored using the International Caries Detection and Assessment System (ICDAS). Saliva is collected for physiological measures. Salivary Deoxyribose Nucleic Acid (DNA) is extracted for genetic studies including epigenetics using the SeqCap Epi Enrichment Kit. Targets of interest are being confirmed by pyrosequencing to identify potential epigenetic markers of caries risk.

Discussion

This study will examine a wide range of potential determinants for childhood dental caries and evaluate inter-relationships amongst them. The findings will provide an evidence base to plan and implement improved preventive strategies.