Meta-analysis prediction intervals are under reported in sport and exercise medicine.

AIM: Prediction intervals are a useful measure of uncertainty for meta-analyses that capture the likely effect size of a new (similar) study based on the included studies. In comparison, confidence intervals reflect the uncertainty around the point estimate but provide an incomplete summary of the underlying heterogeneity in the meta-analysis. This study aimed to estimate (i) the proportion of meta-analysis studies that report a prediction interval in sports medicine; and (ii) the proportion of studies with a discrepancy between the reported confidence interval and a calculated prediction interval. METHODS: We screened, at random, 1500 meta-analysis studies published between 2012 and 2022 in highly ranked sports medicine and medical journals. Articles that used a random effect meta-analysis model were included in the study. We randomly selected one meta-analysis from each article to extract data from, which included the number of estimates, the pooled effect, and the confidence and prediction interval. RESULTS: Of the 1500 articles screened, 866 (514 from sports medicine) used a random effect model. The probability of a prediction interval being reported in sports medicine was 1.7% (95% CI = 0.9%, 3.3%). In medicine the probability was 3.9% (95% CI = 2.4%, 6.6%). A prediction interval was able to be calculated for 220 sports medicine studies. For 60% of these studies, there was a discrepancy in study findings between the reported confidence interval and the calculated prediction interval. Prediction intervals were 3.4 times wider than confidence intervals. CONCLUSION: Very few meta-analyses report prediction intervals and hence are prone to missing the impact of between-study heterogeneity on the overall conclusions. The widespread misinterpretation of random effect meta-analyses could mean that potentially harmful treatments, or those lacking a sufficient evidence base, are being used in practice. Authors, reviewers, and editors should be aware of the importance of prediction intervals.

Being Bayesian in the 2020s: opportunities and challenges in the practice of modern applied Bayesian statistics.

Building on a strong foundation of philosophy, theory, methods and computation over the past three decades, Bayesian approaches are now an integral part of the toolkit for most statisticians and data scientists. Whether they are dedicated Bayesians or opportunistic users, applied professionals can now reap many of the benefits afforded by the Bayesian paradigm. In this paper, we touch on six modern opportunities and challenges in applied Bayesian statistics: intelligent data collection, new data sources, federated analysis, inference for implicit models, model transfer and purposeful software products. This article is part of the theme issue 'Bayesian inference: challenges, perspectives, and prospects'.

Healthcare services for people with acquired disability in South-East Queensland, Australia: Assessing potential proximity and its association with service obstacles.

This study described access potential in South-East Queensland, to healthcare services commonly used by people with acquired disability; and investigated the association between service proximity and perceived service obstacles. First, we described accessibility by conducting a spatial analysis to create maps of potential accessibility to health services in South-East Queensland. Queensland statistical area level 2 (SA2) locations were combined with the residential locations of participants from a longitudinal cohort study involving people with ABI and SCI. The locations of selected health services of interest were identified from Health Direct's National Health Service Directory. Travel times via motor vehicle were modelled with Robust Gaussian Stochastic Process, to smoothly interpolate between the sparse time-to-service observations. Probabilistic predicted values were generated from the model and were used to construct service accessibility maps of South-East Queensland. Disability population data were used to identify SA2s with relatively low service access but a high disability population. Second, we examined perceived service obstacles, by investigating the relationship between potential access to services and perceived service obstacles was examined using data from 63 people with ABI or SCI discharged from the specialist state-wide rehabilitation services, located in South-East Queensland. Obstacles to accessing service in relation to resource availability, transportation and finances were collected three-months after discharge, using the Service Obstacle Scale. Travel times to the closest health service were computed for each individual and were compared to their Service Obstacle Scale responses. Access potential was highly variable, particularly for allied health services. We identified several low-access, high-disability population areas. These hotpots of poor access were generally to the north and west of greater Brisbane. Longer travel times to allied health services were associated with 260% higher odds of agreeing that resource availability was an obstacle to accessing services. Policy makers should be concerned with the hotspots of poor access identified.

The influence of personal factors, unmet need and service obstacles on the relationship between health service use and outcome after brain injury.

BACKGROUND: This exploratory study aimed to: (i) examine the relationship between health service use and quality of life, psychological wellbeing, global function and participation after discharge from brain injury inpatient rehabilitation, and (ii) determine the influence of personal factors, unmet need for services and service obstacles on the relationship between service use and these outcomes. METHODS: Using a prospective cohort design, 41 adults with acquired brain injury (median age = 46 years; 71% male; 61% severe traumatic injury) were followed for 6-months after discharge from specialist brain injury inpatient rehabilitation. Service use was continuously recorded and obtained through data linkage methods, focusing on the use of: outpatient medical services, outpatient nursing, outpatient allied health; medical acute services; incidents of re-hospitalization; and transitional rehabilitation service use. Outcome questionnaire measures were completed via telephone, at 6-months after discharge, and included: the EuroQol-5D; Depression Anxiety and Stress Scale, Mayo-Portland Adaptability Inventory and Sydney Psychosocial Reintegration Scale. Data were analyzed in a heterogeneous treatment effects framework, using Bayesian Additive Regression Trees. RESULTS: There was weak evidence that transitional rehabilitation service use was associated with better psychological wellbeing scores. The posterior probability of lower depression, anxiety and stress scores was .87, .81 and .86, respectively (average treatment effect). There was also weak evidence that re-hospitalization was associated with worse independent living skills scores. The posterior probability of worse scores was .87. However, most re-hospitalizations were due to unavoidable medical complications. We did not find that place of residence at discharge, marital status, unmet need, or service obstacles affected the relationship between service use and the studied outcomes. CONCLUSIONS: This study may highlight the importance of participation in transitional rehabilitation, in the 6-months after discharge from brain injury rehabilitation. Replication in a larger sample size is required to confirm these findings.

Peer Presence Increases Session Ratings of Perceived Exertion.

PURPOSE: This study aimed to examine the effect of peer presence on session rating of perceived exertion (RPE) responses. METHOD: Fourteen males, with mean (SD) age 22.4 (3.9) years, peak oxygen uptake 48.0 (6.6) mL·kg-1·min-1, and peak power output 330 (44) W, completed an incremental cycling test and 3 identical experimental sessions, in groups of 4 or 5. Experimental sessions involved 24 minutes of cycling, whereby the work rate alternated between 40% and 70% peak power output every 3 minutes. During cycling, heart rate was collected every 3 minutes, and session-RPE was recorded 10 minutes after cycling, in 3 communication contexts: in written form unaccompanied (intrapersonal communication), verbally by the researcher only (interpersonal communication), and in the presence of the training group. Session-RPE was analyzed using ordinal regression and heart rate using a linear mixed-effects model, with models fit in a Bayesian framework. RESULTS: Session-RPE was voted higher when collected in the group's presence compared with when written (odds ratio = 4.26, 95% credible interval = 1.27-14.73). On average, the posterior probability that session-RPE was higher in the group setting than when written was .53. Session-RPE was not different between the group and verbal, or verbal and written collection contexts. CONCLUSIONS: This study suggests that contextual psychosocial inputs influence session-RPE and highlights the importance of session-RPE users controlling the measurement environment when collecting votes.

Female (Under) Representation in Exercise Thermoregulation Research.

BACKGROUND: Despite an increasing rate of women participating in professional sports, emergency services, and military settings where they are exposed to exertional heat stress, our understanding of female thermoregulation and the detrimental effects of heat on women's performance, especially regarding the menstrual cycle, is limited. This review aimed to quantify the representation of women in exercise thermoregulation research between 2010 and 2019 and the frequency that these articles reported details pertaining to female participants' menstrual cycle to determine the volume of novel research that is directly relevant to this growing population. METHODS: Original exercise thermoregulatory studies published in three major sports medicine databases (PubMed, MEDLINE, and SPORTDiscus) between 2010 and 2019 were surveyed. Articles were screened to determine the number of female and male participants in the study and whether studies involving women reported menstrual orientation or phase. Research involving healthy adult participants and an exercise protocol with a thermoregulatory outcome measure were included in the review. RESULTS: A total of 1407 articles were included in the review, involving 28,030 participants. The annual representation of women ranged from a mean of 11.6% [95% credible interval (CI); 9.2, 14.3] to 17.8% [95% CI; 15.2, 20.6] across the 10 years, indicating studies predominantly included men. Nonetheless, there was a small statistical increase in the overall proportion of women, with a mean overall proportion change of 0.7% [95% CI; 0.2, 1.2] per year. The increase appeared to be driven by a reduction in the number of studies including only men, rather than studies including more women alongside men, or increased women-only studies. Less than one third of articles involving women reported the menstrual orientation of participants and less than one quarter reported both menstrual orientation and phase. This study shows that women were proportionally underrepresented in exercise thermoregulation research during the past decade and the majority of studies did not report menstrual cycle details of female participants. Researchers should consider including women in future work where their inclusion could contribute meaningful data that enhance the evidence-based and ultimately improves our comprehension of women's thermal physiology.

Including mobile-only telephone users in a statewide preventive health survey-Differences in the prevalence of health risk factors and impact on trends.

The Queensland preventive health survey is conducted annually to monitor the prevalence of behavioural risk factors in the north-east Australian state. Prompted by domestic and international trends in mobile telephone usage, the 2015 survey incorporated both mobile and landline telephone numbers from a list-based sampling frame. Estimates for landline-accessible and mobile-only respondents are compared to assess potential bias in landline-only surveys in the context of public health surveillance. Significant differences were found in subcategories of all health prevalence estimates considered (alcohol consumption, body mass index, smoking, and physical activity) from 2015 survey results. Results from Australian and international studies that have considered mobile telephone non-coverage bias are also summarised and discussed. We find that adjusting for sampling biases of telephone surveys by weighting does not fully compensate for the differences in prevalence estimates. However, predicted trends from previous years' surveys only differ significantly for the 2015 prevalence estimates of alcohol consumption. We conclude that the inclusion of mobile telephones into standard telephones surveys is important for obtaining valid, reliable and representative data to reduce bias in health prevalence estimates. Importantly, unlike some international experiences, the addition of mobiles telephones into the Queensland preventive health survey occurred before population trends were significantly affected.