Online labor markets and the humans that power them serve a critical role in the advancement of artificial intelligence and supervised machine learning via the creation of useful training datasets. ...The use of human effort in online labor markets is not enough, however, as a key factor is understanding the possible interventions that market operators can leverage to incentivize human effort among their labor force. We propose that platforms could implement mechanisms such as rewards or punishments at individual or group levels to incentivize real-effort and output. We apply our interventions using a collaborative image tagging experiment—a folksonomy—and the results provide interesting insights and nonobvious consequences. On average, interventions applied at the group level outperformed interventions applied at the individual level. Punishing the group provided the most controversial incentive strategy and provided a nonobvious significant improvement in effort. Rewarding or sanctioning an individual had similar effects on average, with both treatments leading to significant increases in effort post-intervention. In contrast to predictions, sanctioning appears to have significantly motivated those that were punished. Overall, the interventions applied in our real-effort collaborative image tagging experiment had a significant impact on behavior, which provides guidance for online labor market operators and the use of incentives in the creation of labeled machine learning training datasets.
The impact of multi-channel technology-enabled digital goods on the sales of the physical counterpart faces uncertainty in the electronic commerce domain. We address the issue empirically by ...identifying the effect of the availability of digitally-delivered movies on physical DVD movie sales. Unique to our study is our interest in not only purchased digital goods but rented digital goods as well. We construct a robust panel dataset consisting of movie data collected from Amazon and Barnes and Noble on the same day for every movie observed. A key feature of our dataset is the multi-channel availability of digital purchase and digital rental movie formats at Amazon. Our results show that the availability of the digital purchase format does not have a significant effect on DVD sales. Surprisingly, the availability of the digital rental format is associated with a significant reduction in DVD sales. The results imply that a product substitution effect may be occurring between the digital rental and the physical DVD purchase of the same movie. We conduct robustness tests to show under which conditions the effect is greatest. Our results also provide practical implications to inform strategies regarding movie format release windows.
•Digital rentals significantly cannibalize sales of physical movies.•Digital purchases do not significantly cannibalize sales of physical movies.•Cannibalization is driven by new movies that have few reviews and large discounts.•Cannibalization diminishes after a physical movie has been released for 2000 days.
Conservative therapies are recommended as initial treatment for male lower urinary tract symptoms. However, there is a lack of evidence on effectiveness and uncertainty regarding approaches to ...delivery.
The objective was to determine whether or not a standardised and manualised care intervention delivered in primary care achieves superior symptomatic outcome for lower urinary tract symptoms to usual care.
This was a two-arm cluster randomised controlled trial.
The trial was set in 30 NHS general practice sites in England.
Participants were adult men (aged ≥ 18 years) with bothersome lower urinary tract symptoms.
Sites were randomised 1 : 1 to deliver the TReatIng Urinary symptoms in Men in Primary Health care using non-pharmacological and non-surgical interventions trial intervention or usual care to all participants. The TReatIng Urinary symptoms in Men in Primary Health care using non-pharmacological and non-surgical interventions intervention comprised a standardised advice booklet developed for the trial from the British Association of Urological Surgeons' patient information sheets, with patient and expert input. Patients were directed to relevant sections by general practice or research nurses/healthcare assistants following urinary symptom assessment, providing the manualised element. The healthcare professional provided follow-up contacts over 12 weeks to support adherence to the intervention.
The primary outcome was the validated patient-reported International Prostate Symptom Score 12 months post consent. Rather than the minimal clinically important difference of 3.0 points for overall International Prostate Symptom Score, the sample size aimed to detect a difference of 2.0 points, owing to the recognised clinical impact of individual symptoms.
A total of 1077 men consented to the study: 524 in sites randomised to the intervention arm (
= 17) and 553 in sites randomised to the control arm (
= 13). A difference in mean International Prostate Symptom Score at 12 months was found (adjusted mean difference of -1.81 points, 95% confidence interval -2.66 to -0.95 points), with a lower score in the intervention arm, indicating less severe symptoms. Secondary outcomes of patient-reported urinary symptoms, quality of life specific to lower urinary tract symptoms and perception of lower urinary tract symptoms all showed evidence of a difference between the arms favouring the intervention. No difference was seen between the arms in the proportion of urology referrals or adverse events. In qualitative interviews, participants welcomed the intervention, describing positive effects on their symptoms, as well as on their understanding of conservative care and their attitude towards the experience of lower urinary tract symptoms. The interviews highlighted that structured, in-depth self-management is insufficiently embedded within general practitioner consultations. From an NHS perspective, mean costs and quality-adjusted life-years were similar between trial arms. The intervention arm had slightly lower mean costs (adjusted mean difference of -£29.99, 95% confidence interval -£109.84 to £22.63) than the usual-care arm, and a small gain in quality-adjusted life-years (adjusted mean difference of 0.001, 95% confidence interval -0.011 to 0.014).
The intervention showed a small, sustained benefit for men's lower urinary tract symptoms and quality of life across a range of outcome measures in a UK primary care setting. Qualitative data showed that men highly valued the intervention. Intervention costs were marginally lower than usual-care costs. Limitations of the study included that trial participants were unmasked, with limited diversity in ethnicity and deprivation level. Additional research is needed to assess the applicability of the intervention for a more ethnically diverse population..
This trial is registered as ISRCTN11669964.
This award was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme (NIHR award ref: 16/90/03) and is published in full in
; Vol. 28, No. 13. See the NIHR Funding and Awards website for further award information.
The adoption of health IT transforms communication between care providers and patients. Unfortunately, research on the transformation of communication has produced conflicting results, creating ...tension regarding its efficacy among healthcare professionals. In this paper, we propose that nurse and physician communication performance mediate the relationship between health IT implementation and patient outcomes. We test the mediating role of communication with a hospital-level data set spanning 2011 through 2015. The specific health information technologies we investigate include EMR documentation, computerized physician order entry (CPOE) systems, clinical decision support (CDS) systems, and health information exchanges (HIE). Our results provide that EMR documentation, CPOE, and HIE directly improve communication between care providers and patients as well as patient outcomes. Further, nurse-patient and physician-patient communication mediates the relationship between health IT implementation and patient outcomes. The mediating effect extends the positive benefits to patient outcomes following technology implementation. We also find that poor communication with patients directly increases mortality, decreases satisfaction, and decreases loyalty. Surprisingly, CDS has a negative relationship on communication and patient outcomes. Our findings contribute to the information systems and healthcare literatures by demonstrating the need to account for the multidimensional nature of healthcare and by providing context for the positive and negative effects previously discovered. Furthermore, the results offer practical and theoretical implications for leveraging specific health IT adoption and for realigning federal incentive structures for hospitals.
•Effects of healthcare technologies on outcomes are partially mediated by improved communication with physicians and nurses.•Electronic documentation and information exchanges have the most profound impacts on communication and patient outcomes.•Decision support systems negatively affect patient outcomes by worsening nurse-patient and physician-patient communication.
Piracy is a significant source of concern facing software developers, music labels, and movie production companies, to name a few. Digital goods producers and government entities argue that there are ...victims of piracy, whereas pirates may perceive their actions to be victimless. Regarding implications of our research, we extend the theory of planned behavior (TPB) by theorizing that attitudes, subjective norms, and perceived behavioral control could influence perceptions of moral obligations as a consequence of the desire to rationalize unethical behavior. Unlike prior literature, we manipulate the rationalization of moral obligations due to the victimless view toward piracy and show how moral obligations become important determinants of piracy behavior. Accordingly, our demonstrated malleability of morals may be an important path through which individuals are able to continue past behaviors. We also conduct a second study to identify the effect of implementing an educational message from a fictitious software company to exogenously nudge the pirate and influence the impact of perceived moral obligations on intentions to pirate. Our results show that the introduction of an exogenous educational message is an effective piracy mitigation strategy.
We investigate changes to the value that individuals place on the online disclosure of their private information in the presence of multiple privacy factors. We capture individuals’ ...willingness-to-accept for a privacy disclosure in a series of randomized experiments that manipulate characteristics of a required privacy disclosure by altering the information context, the intended secondary use of the disclosed private information, and the requirement to disclose personally identifying information. We collect data from two populations (college students and Amazon Mechanical Turk workers) to aid with generalizability of our results. Across the experiments, we consistently observe null effects for each of the privacy factors. The results provide a unique perspective on privacy valuations by showing that results from prior research on simple privacy decisions may not translate to more realistic, complex privacy disclosure decisions that involve multiple factors. Our findings suggest that disclosing private information may be an all or nothing type of decision as opposed to an activation of individual factors proposed by prior literature as important in a multidimension private information disclosure. This study provides managerial insight into the possible evolution of online disclosure decisions, especially in settings that incorporate multiple disclosure dimensions.
We investigate changes to the value that individuals place on the online disclosure of their private information in the presence of multiple privacy factors. We use an incentive-compatible mechanism to capture individuals’ willingness-to-accept (WTA) for a privacy disclosure in a series of three randomized experiments. Each experiment manipulates characteristics of a required privacy disclosure by altering the information context, the intended secondary use of the disclosed private information, and the requirement to disclose personally identifying information. We collect data from two populations (college students and Amazon Mechanical Turk workers) to aid with generalizability of our results. As methodological checks to rule out lack of awareness in the participants, we first increase the saliency of the privacy disclosure characteristics in the second experiment and then require participants to watch a video on the potential consequences of disclosing private information in the third experiment. Across the three experiments, we consistently observe null effects for each of the privacy factors, with two population-dependent exceptions in the second study. Our participants do acknowledge the increased risk introduced by the experimental factors, and the increased saliency and awareness do lead to higher privacy valuations on average. However, there is no consistent manifestation as significant main effects for the three privacy factors. This is in contrast to prior research, which has found significant effects for each of these factors when studied separately. The results provide a unique perspective on privacy valuations by showing that results from prior research on simple privacy decisions may not translate to more realistic, complex privacy disclosure decisions that involve multiple factors.
The online appendix is available at
https://doi.org/10.1287/isre.2018.0818
.
ObjectivesTo estimate the cost-effectiveness of a primary care intervention for male lower urinary tract symptoms (LUTS) compared with usual care.DesignEconomic evaluation alongside a cluster ...randomised controlled trial from a UK National Health Service (NHS) perspective with a 12-month time horizon.SettingThirty NHS general practice sites in England.Participants1077 men aged 18 or older identified in primary care with bothersome LUTS.InterventionsA standardised and manualised intervention for the treatment of bothersome LUTS was compared with usual care. The intervention group (n=524) received a standardised information booklet with guidance on conservative treatment for LUTS, urinary symptom assessment and follow-up contacts for 12 weeks. The usual care group (n=553) followed local guidelines between general practice sites.MeasuresResource use was obtained from electronic health records, trial staff and participants, and valued using UK reference costs. Quality-adjusted life-years (QALYs) were calculated from the EQ-5D-5L questionnaire. Adjusted mean differences in costs and QALYs and incremental net monetary benefit were estimated.Results866 of 1077 (80.4%) participants had complete data and were included in the base-case analysis. Over the 12-month follow-up period, intervention and usual care arms had similar mean adjusted costs and QALYs. Mean differences were lower in the intervention arm for adjusted costs −£29.99 (95% CI −£109.84 to £22.63) while higher in the intervention arm for adjusted QALYs 0.001 (95% CI −0.011 to 0.014). The incremental net monetary benefit statistic was £48.01 (95% CI −£225.83 to £321.85) at the National Institute for Health and Care Excellence UK threshold of £20 000 per QALY. The cost-effectiveness acceptability curve showed a 63% probability of the intervention arm being cost-effective at this threshold.ConclusionsCosts and QALYs were similar between the two arms at 12 months follow-up. This indicates that the intervention can be implemented in general practice at neutral cost.Trial registration numberISRCTN11669964.
There are many contexts in which an “everybody else is doing it” attitude is relevant. We evaluate the impact of this attitude in a multithreshold public goods game. We use a lab experiment to study ...the role of providing information about contribution behavior to targeted subsets of individuals, and its effect on coordination. Treatments include one in which no information is provided and three other treatments, i.e., where information is provided to a random sample of subjects; to those whose contributions are below the average of their group, and to those whose contributions are above the average of their group. We find that the random provision of information is no different than not providing information. More important, average contributions improve with targeted treatments. Coordination waste is also lower with targeted treatments. The insights from this research are more broadly relevant in the contexts of piracy, open innovation, and crowdfunding.
The online appendix is available at
https://doi.org/10.1287/isre.2016.0663
Lower urinary tract symptoms (LUTS) can relate to urinary storage or voiding. In men, the prevalence and severity of LUTS increases with age, with a significant impact on quality of life. The ...majority of men presenting with LUTS are managed by their general practitioner (GP) in the first instance, with conservative therapies recommended as the initial treatment. However, the provision of conservative therapies in primary care is variable and can be time and resource limited. GPs require practical resources to enhance patient engagement with such interventions. TRIUMPH aims to determine whether a standardised and manualised care intervention delivered in primary care achieves superior symptomatic outcome for LUTS versus usual care.
TRIUMPH is a two-arm, cluster randomised controlled trial (RCT) being conducted in 30 National Health Service (NHS) general practices in England. The TRIUMPH intervention comprises a standardised LUTS advice booklet developed for the trial with patient and healthcare professional (HCP) consultation. The booklet is delivered to patients by nurses/healthcare assistants following assessment of their urinary symptoms. Patients are directed to relevant sections of the booklet, providing the manualised element of the intervention. To encourage adherence, HCPs provide follow-up contacts over 12 weeks. Practices are randomised 1:1 to either deliver the TRIUMPH intervention or a usual care pathway. The patient-reported International Prostate Symptom Score (IPSS) at 12 months post consent is the primary outcome. Secondary outcomes include cost-effectiveness, patient-reported outcomes on LUTS, quality of life, and patient and HCP acceptability and experience of the intervention. Primary analyses will be conducted on an intention-to-treat basis.
It is unclear whether conservative therapies for male LUTS are effectively delivered in primary care using current approaches. This can lead to men being inappropriately referred to secondary care or experiencing persistent symptoms. Primary care, therefore, holds the key to effective treatment for these men. The TRIUMPH intervention, through its standardised and manualised approach, has been developed to support GP practices in delivering effective conservative care. This pragmatic, cluster RCT should provide robust evidence in a primary-care setting to inform future guidelines.
ISCRTN registry, ID: ISRCTN11669964 . Registered on 12 April 2018.
PDF
