Aggregated Sales Data Research Articles

Antibacterial resistance and the cost of affecting demand: The case of UK antibiotics

Consumption of broad-spectrum antibiotics is associated with rising antimicrobial resistance (AMR) levels. The use of broad-spectrum drugs, particularly of cephalosporins, quinolones, and co-amoxiclav contributes the most to the rise in AMR. We use aggregate sales data on antibiotics from the UK to estimate structural demand models and reveal drug substitution patterns. We then simulate alternative tax schemes to evaluate the effectiveness of shifting demand from broad- to narrow-spectrum drugs. Our estimates suggest that these policies can be highly effective in demand management and come at a relatively low cost regarding changes in consumer and producer surplus.

A Random Coefficients Model of Seafood Demand: Implications for Consumer Preferences and Substitution Patterns

Discrete choice models of demand are growing in popularity for understanding markets for seafood, but have thus far been limited to applications using individual-level choice data. The random coefficients logit model is a discrete choice demand model designed for aggregate sales data and imparts a number of theoretical and empirical advantages. Instrumental variables account for price endogeneity, which can arise when there are unobserved product characteristics. Furthermore, correlated preferences can be accommodated in the random coefficients as well as through demographic interactions, which is especially important for seafood where product characteristics are primarily qualitative. We estimate this model for salmon fillets using four years of county-level seafood sales in California, and demonstrate the insights that can be drawn regarding consumer preferences and substitution patterns. Although the model is computationally burdensome, it offers considerable potential for further seafood demand analysis.

Residential green space and mental health-related prescription medication sales: An ecological study in Belgium

BackgroundResidential green space has been associated with mental health benefits, but how such associations vary with green space types is insufficiently known. ObjectiveWe aimed to investigate associations between types and quantities of green space and sales of mood disorder medication in Belgium. MethodsWe used aggregated sales data of psycholeptics and psychoanaleptics prescribed to adults from 2006 to 2014. Generalized mixed effects models were used to investigate associations between relative covers of woodland, low-green, grassland, and garden, and average annual medication sales. Models were adjusted for socio-economic background variables, urban-rural differences, and administrative region, and included random effects of latitude and longitude. ResultsUrban census tracts were associated with 9–10% higher medication sales. In nationwide models, a 10% increase in relative cover of woodland, garden, and grass was associated with a 1–2% decrease in medication sales. The same association was found for low green but only for men. In stratified models, a 10% increase in relative cover of any green space type in urban census tracts was associated with a decrease of medication sales by 1–3%. In rural census tracts, no protective associations between green space and mood disorder medication sales were observed, with the exception of relative woodland cover for women (−1%), and low green was associated with higher medication sales (+6–7%). ConclusionsTaken together, these results suggest that living in green environments may be beneficial for adult mental health. Woodland exposure seemed the most beneficial, but the amount of green space was more important than the type. Results underline the importance of conserving green space in our living environment, for the conservation of biodiversity and for human health.

Changes in the calorie and nutrient content of purchased fast food meals after calorie menu labeling: A natural experiment.

BackgroundCalorie menu labeling is a policy that requires food establishments to post the calories on menu offerings to encourage healthy food choice. Calorie labeling has been implemented in the United States since May 2018 per the Affordable Care Act, but to the best of our knowledge, no studies have evaluated the relationship between calorie labeling and meal purchases since nationwide implementation of this policy. Our objective was to investigate the relationship between calorie labeling and the calorie and nutrient content of purchased meals after a fast food franchise began labeling in April 2017, prior to the required nationwide implementation, and after nationwide implementation of labeling in May 2018, when all large US chain restaurants were required to label their menus.Methods and findingsWe obtained weekly aggregated sales data from 104 restaurants that are part of a fast food franchise for 3 national chains in 3 US states: Louisiana, Mississippi, and Texas. The franchise provided all sales data from April 2015 until April 2019. The franchise labeled menus in April 2017, 1 year prior to the required nationwide implementation date of May 2018 set by the US Food and Drug Administration. We obtained nutrition information for items sold (calories, fat, carbohydrates, protein, saturated fat, sugar, dietary fiber, and sodium) from Menustat, a publicly available database with nutrition information for items offered at the top revenue-generating US restaurant chains. We used an interrupted time series to find level and trend changes in mean weekly calorie and nutrient content per transaction after franchise and nationwide labeling. The analytic sample represented 331,776,445 items purchased across 67,112,342 transactions. Franchise labeling was associated with a level change of −54 calories/transaction (95% confidence interval [CI]: −67, −42, p < 0.0001) and a subsequent 3.3 calories/transaction increase per 4-week period (95% CI: 2.5, 4.1, p < 0.0001). Nationwide implementation was associated with a level decrease of −82 calories/transaction (95% CI: −88, −76, p < 0.0001) and a subsequent −2.1 calories/transaction decrease per 4-week period (95% CI: −2.9, −1.3, p < 0.0001). At the end of the study, the model-based predicted mean calories/transaction was 4.7% lower (change = −73 calories/transaction, 95% CI: −81, −65), and nutrients/transaction ranged from 1.8% lower (saturated fat) to 7.0% lower (sugar) than what we would expect had labeling not been implemented. The main limitations were potential residual time-varying confounding and lack of individual-level transaction data.ConclusionsIn this study, we observed that calorie labeling was associated with small decreases in mean calorie and nutrient content of fast food meals 2 years after franchise labeling and nearly 1 year after implementation of labeling nationwide. These changes imply that calorie labeling was associated with small improvements in purchased meal quality in US chain restaurants.

Favorable Commercial and Health Behavior Impacts of a Healthy Vending Policy at an Australian University

BackgroundThe World Health Organization recognizes universities as an important health-promotion setting, including in healthy food provision. Previous research shows that healthy food retail interventions also need to consider commercial sustainability, including financial outcomes, and should take a holistic approach to consumer experience. ObjectiveOur aim was to determine the health behavior and commercial outcomes of a multicomponent traffic light–based healthy vending policy implemented as one part of a holistic university food policy. The hypothesis was that purchases of less healthy “red” beverages would decrease compared with predicted sales, that purchases of healthier “green” and “amber” alternatives would increase, and that there would be no change in revenue. DesignA quasi-experimental design evaluated a real-world food policy using monthly aggregated sales data to compare pre-intervention (January 2016 to March 2018) and post-intervention period sales (December 2018 to December 2019). Participants/settingElectronic sales data were collected from 51 beverage vending machines across 4 university campuses in Victoria, Australia. InterventionA multicomponent policy was implemented between April and November 2018. Beverages were classified using a voluntary state government traffic light framework. Policy included display ≤20% red beverages and ≥50% green beverages; machine traffic light labeling; health-promoting machine branding; review of machine placement; and recycled bottle packaging. Main outcome measuresChanges in red, amber, and green volume sales, and revenue compared with predicated sales. Statistical analyses performedInterrupted time series analysis of sales data compared post-policy sales with predicted sales. ResultsIn the 13th month post-policy implementation, there was a 93.2% (95% CI +35.9% to +150.5%) increase in total beverage volume sold and an 88.6% (95% CI +39.2% to +138.1%) increase in revenue. There was no change in red beverage volume sold, but increases in green (+120.8%; 95% CI +59.0% to +182.6%) and amber (+223.2%; 95% CI +122.4% to +323.9%) volume sold. ConclusionsSustained behavior change and commercial outcomes suggest that holistic vending interventions can effectively promote healthier beverage sales.

Decoupling the individual effects of multiple marketing channels with state space models

An important problem in marketing is understanding the impact of various marketing efforts on sales and revenue. If specific measures have not been taken before hand to distinguish responders across the various marketing initiatives, it becomes increasingly difficult to assess the effectiveness of investment across individual marketing channels. We present State Space model to estimate the individual channel effects using aggregate sales or response data across all channels. The proposed framework allows for varying carry over effects across marketing channels. Also, the proposed framework allows for differing rates of decay across marketing channels. We demonstrate its use when data on sales due to individual marketing channels is not available and only aggregated sales data are available. The proposed State Space modeling approach offers the advantage of: (1) allowing for varying rates of decay across marketing channels, and (2) allowing for a natural way of modeling time series dynamics. The approach also opens the way for more comprehensive marketing-mix optimization by allowing varying rates of decay.

Estimating Complementarity With Large Choice Sets: An Application to Mergers

Standard discrete choice demand models assume that products are substitutes. Merger analyses based on these models may overstate consumer harm when producers of complementary products merge. Allowing for demand complementarity greatly complicates demand estimation, particularly when the number of choices is large. We introduce a straightforward Generalized Method of Moments estimator that identifies preferences allowing for (1) potential consumption complementarity, (2) price endogeneity and (3) large choice sets. Our estimator parsimoniously leverages information on consumer level bundle specific purchases and aggregate sales data. We apply this estimator to the chips and soda market and find a high degree of complementarity between these product groups. We show that a merger between PepsiCo/Frito-Lay and Dr. Pepper would increase soda prices by 30% less than suggested by a model that does not account for complementarity. Post-merger chip prices decrease. Overall, accounting for complementarity leads to positive welfare gains in some markets with large numbers of Frito-Lay varieties.

Estimating the effect of calorie menu labeling on calories purchased in a large restaurant franchise in the southern United States: quasi-experimental study

ObjectiveTo evaluate whether calorie labeling of menus in large restaurant chains was associated with a change in mean calories purchased per transaction.DesignQuasi-experimental longitudinal study.SettingLarge franchise of a national fast food...

Collusion and Information Exchange

Antitrust authorities view that exchange of individual firms’ sales data is more anti-competitive than that of aggregate sales data. In this paper, I survey antitrust implications of such inter-firm information exchange. I argue that both types of information exchange are anti-competitive under some circumstances. More precisely, I compare profits when each type of information exchange is allowed to that when firms can only observe their own sales (Stigler’s secret price-cutting model), and the former is bigger than the latter. I also provide a general method to bound the equilibrium profits without such information exchange.

Examining the impact of celebrity endorsements across consumer segments: an empirical study of Tiger Woods’ endorsement effect on golf equipment

This paper empirically examines whether endorsement effects differ across consumer types–expert vs. novice. The study uses aggregate sales data from the golf equipment industry covering two unique retail channels (on-course and off-course) to estimate endorsement effects for Tiger Woods. The study finds that Tiger Woods’ endorsement of Titleist woods had a greater effect in the off-course retail channel frequented by novice golfers than the on-course channel frequented by experts. I determine that such an effect led to over 28 thousand more Titleist woods being sold and an additional $9.2 million in revenue for the 9 months prior to Tiger Woods switching to endorse Nike apparel and equipment. I also determine that Tiger Woods’ endorsement led to over $500,000 in additional profit for each of the last 5 months of his endorsement contract with Titleist, which provided a 49% Return on Investment from US sales of Titleist woods, putters, and irons. Lastly, I present preliminary evidence to explain a possible reason why the observed endorsement effects varied by consumer types—Tiger Woods’ endorsements were informative to consumers. Specifically, I find support indicating Tiger Woods’ endorsement was more informative to novice golfers than experts.

Measuring the Bullwhip Effect: Discrepancy and Alignment Between Information and Material Flows

The bullwhip effect is a phenomenon commonly observed in supply chains. It describes how demand variance amplifies from a downstream site to an upstream site due to demand information distortion. Two different bullwhip effect measures have been used in the literature. Theorists analyze the bullwhip effect based on the information flow (i.e., order and demand information), whereas most empiricists measure it according to the material flow (i.e., shipment and sales data). It is unclear how much the discrepancy between these two measures is, and, if significant, how to reconcile the discrepancy. In this paper, we illustrate and quantify the discrepancy under three inventory systems. For the system with stationary demand and ample supply, we show that the bullwhip effect measure based on the material-flow data is always greater than that based on the information flow. For the system with correlated demand and for the system with supply shortages, we derive conditions under which the material flow measure is either greater or less than the information flow measure. We find that the discrepancy is driven by four factors: stocking level, lead time, demand correlation, and supply service level. We further propose a method to reduce the discrepancy by using the sample variances of aggregated sales data. Our method works for common demand processes with short-range dependence, and it does not require the knowledge of the underlying base-stock levels.

Understanding the 4As of Social Marketing for HIV Prevention in India

Social marketing has gained popularity in health used to achieve the “social good.” However, there is lack of research beyond the aggregated sales data exploring the debate on socioeconomic and cultural relevance of such programs in public health in a low middle income context. The purpose of this article is therefore (a) to understand the importance of socioeconomic context that moderate acceptability, affordability, accessibility, and availability (4As) of social marketing for HIV/AIDS prevention in India and (b) analyze the barriers they may pose during the operationalization of the programs. This article performed theory-driven evaluation using qualitative research tools like documentary analysis and in-depth interviews of relevant stakeholders. Thematic analysis was followed to classify the emergent data. The 4As comprising “acceptability,” “affordability,” “availability,” and “accessibility” proposed in this article encapsulate the important role played by socioeconomic and cultural dimensions in shaping the social marketing programs for HIV/AIDS prevention in India. Social marketing for “whom,” “at what cost,” and “under what circumstances” emerge as an interesting debate in this article. An enhanced understanding of how social marketing harnesses the prevalent sociocultural and economic dynamics in a heterogeneous context like India not only refines social marketing as a theory but also provides opportunities for effective health policy making resulting in better health outcomes.

Trends and patterns of antibiotic consumption in Shanghai municipality, China: a 6 year surveillance with sales records, 2009-14.

The objective of this study was to identify the trends and patterns of total antibiotic use in Shanghai from 2009 to 2014. Sales records were collected from a minimum of 160 hospitals and 241 primary healthcare settings and used as a proxy for consumption. Antibiotic sales expressed in DDD per 1000 inhabitants per day (DID) were calculated. We compared patterns of antibiotic use in the Shanghai municipality, populated by 24 million inhabitants, with European countries using indicators from the European Surveillance of Antimicrobial Consumption (ESAC). Total antibiotic use was highest at 25.9 DID in 2010, fell to 17.8 DID in 2012 and remained stable thereafter. The majority of this reduction occurred over 6 months in 2011. In 2014, two-thirds of the antibiotic sales were to hospitals (serving both inpatients and outpatients) and one-third to primary healthcare institutions. Cephalosporins accounted for 50.2% of total DID, followed by macrolides (18.2%), quinolones (16.0%) and penicillins (7.3%). A fifth of all antibiotics were used in parenteral form. We have successfully used aggregated sales data to monitor antibiotic usage across a large urban population over a 6 year period. A rapid, substantial and sustained reduction in antibiotic usage across the entire health system occurred. This coincided with several interventions across hierarchies in the health sector, including a national campaign. The patterns of antibiotic use indicate persistent preferences for cephalosporins, macrolides, quinolones and parenteral preparations. Further efforts are needed to investigate and improve the quality of antibiotic use.

Utilization of triptans in Sweden; analyses of over the counter and prescription sales.

To enable easier access to triptans, the drug of choice for moderate to severe migraine, some countries have made triptans available without prescription, that is, over the counter (OTC). Concern has been raised about this. The aim of this study was to describe the utilization pattern of triptans in Sweden before and after the OTC switch. Wholesaler and aggregated sales data from all Swedish pharmacies 1991 to 2011 and patient identity data on dispensed prescriptions 2007 and 2011 from the Swedish National Prescribed Drug Register were used to investigate volume and expenditure of triptans. The databases contain complete data for all drugs sold in Sweden or dispensed to all Swedish inhabitants (9.5 million in 2012). Volumes of triptans have increased to 7.0 million defined daily doses (DDD) on prescriptions and 0.7 million DDDs OTC in 2011. Prescriptions were dispensed to 10.0 and 10.1 per 1000 inhabitants in 2007 and 2011, respectively. Although half of those dispensed triptans in 2007 were not in 2011, the incidence remained stable at 2.8 patients per thousand person-years. In 2011, the 10% of the heaviest users accounted for 44% and 48% of dispensed triptans in women and men, respectively. Triptans OTC and the volumes dispensed on prescription have increased as has the DDD per patient purchasing triptans on prescription. However, the number of patient's dispensed triptans on prescription has remained stable. A concern is that almost half of prescribed triptans are purchased by 10% of the users.

Measuring the Bullwhip Effect with Material Flow Data: Biases and Remedies

The bullwhip effect is a phenomenon commonly observed in supply chains. It describes how demand variance amplifies from a downstream site to an upstream site due to demand information distortion. Two different bullwhip effect measures have been used in the literature. Theorists analyze the bullwhip effect based on the information flow (i.e., order and demand information), whereas most empiricists measure it according to the material flow (i.e., shipment and sales data). It is unclear how much the discrepancy between these two measures is, and, if significant, how to reconcile the discrepancy. In this paper, we illustrate and quantify the discrepancy under three inventory systems. For the system with stationary demand and ample supply, we show that the bullwhip effect measure based on the material flow data is always greater than that based on the information flow. For the system with correlated demand and for the system with supply shortages, we derive conditions under which the material-flow measure is either greater or less than the information-flow measure. We find that the discrepancy is driven by four factors: stocking level, lead time, demand correlation, and supply service level. We further propose a method to reduce the discrepancy by using the sample variances of aggregated sales data. Our method works for common demand processes with short-range dependence, and it does not require the knowledge of the underlying base-stock levels.

PP261—Utilization of triptanes in Sweden; analyses of over the counter and prescriptions sales

ABSTRACT Purpose To enable easier access to triptans, the drug of choice for moderate to severe migraine, some countries have made triptansavailable without prescription, that is, over the counter (OTC). Concern has been raised about this. The aim of this study was to describethe utilization pattern of triptans in Sweden before and after the OTC switch.Methods Wholesaler and aggregated sales data from all Swedish pharmacies 1991 to 2011 and patient identity data on dispensed prescrip-tions 2007 and 2011 from the Swedish National Prescribed Drug Register were used to investigate volume and expenditure of triptans. Thedatabases contain complete data for all drugs sold in Sweden or dispensed to all Swedish inhabitants (9.5 million in 2012).Results Volumes oftriptans have increasedto 7.0million defined daily doses (DDD) on prescriptions and 0.7 million DDDs OTC in 2011.Prescriptions were dispensed to 10.0 and 10.1 per 1000 inhabitants in 2007 and 2011, respectively. Although half of those dispensed triptansin 2007 were not in 2011, the incidence remained stable at 2.8 patients per thousand person-years. In 2011, the 10% of the heaviest usersaccounted for 44% and 48% of dispensed triptans in women and men, respectively.Conclusions Triptans OTC and the volumes dispensed on prescription have increased as has the DDD per patient purchasing triptans onprescription. However, the number of patient’s dispensed triptans on prescription has remained stable. A concern is that almost half ofprescribed triptans are purchased by 10% of the users. Copyright © 2014 John Wiley & Sons, Ltd.key words—triptan; migraine; drug utilization; over the counter; prescription; pharmacoepidemiologyReceived 10March2014; Revised 17June2014; Accepted 23June2014

A Closer Look at the Tax Incidence of Instant Lottery Games: An Analysis by Price Point

The tax incidence of different price-point instant lottery games is examined. Theoretical reasons exist for expecting higher-priced instant lottery games to be less regressive than lower-priced instant games. Using game-level data from a sample of states, the empirical results show that higher-priced instant games are significantly less regressive than lower-priced games. For some games, regressivity is rejected in favor of proportionality. In addition, the tax incidence of individual instant games is quite different than that for all instant games combined. This suggests that large differences in individual instant-game tax incidence are masked if aggregated sales data are used.

Forecasting drug utilization and expenditure in a metropolitan health region

BackgroundNew pharmacological therapies are challenging the healthcare systems, and there is an increasing need to assess their therapeutic value in relation to existing alternatives as well as their potential budget impact. Consequently, new models to introduce drugs in healthcare are urgently needed. In the metropolitan health region of Stockholm, Sweden, a model has been developed including early warning (horizon scanning), forecasting of drug utilization and expenditure, critical drug evaluation as well as structured programs for the introduction and follow-up of new drugs. The aim of this paper is to present the forecasting model and the predicted growth in all therapeutic areas in 2010 and 2011.MethodsLinear regression analysis was applied to aggregate sales data on hospital sales and dispensed drugs in ambulatory care, including both reimbursed expenditure and patient co-payment. The linear regression was applied on each pharmacological group based on four observations 2006-2009, and the crude predictions estimated for the coming two years 2010-2011. The crude predictions were then adjusted for factors likely to increase or decrease future utilization and expenditure, such as patent expiries, new drugs to be launched or new guidelines from national bodies or the regional Drug and Therapeutics Committee. The assessment included a close collaboration with clinical, clinical pharmacological and pharmaceutical experts from the regional Drug and Therapeutics Committee.ResultsThe annual increase in total expenditure for prescription and hospital drugs was predicted to be 2.0% in 2010 and 4.0% in 2011. Expenditures will increase in most therapeutic areas, but most predominantly for antineoplastic and immune modulating agents as well as drugs for the nervous system, infectious diseases, and blood and blood-forming organs.ConclusionsThe utilisation and expenditure of drugs is difficult to forecast due to uncertainties about the rate of adoption of new medicines and various ongoing healthcare reforms and activities to improve the quality and efficiency of prescribing. Nevertheless, we believe our model will be valuable as an early warning system to start developing guidance for new drugs including systems to monitor their effectiveness, safety and cost-effectiveness in clinical practice.

Does drinking really decrease in bad times?

This paper investigates the relationship between macroeconomic conditions and drinking using individual-level data from 1987 to 1999 interview years of the “behavioral risk factor surveillance system” (BRFSS). We confirm the procyclical variation in overall drinking identified in previous research using aggregate sales data and show that this largely results from changes in consumption by existing drinkers, rather than movements into or out of drinking. Moreover, the decrease occurring during bad economic times is concentrated among heavy consumers, with light drinking actually rising. We also find no evidence that the decline in overall alcohol use masks a rise for persons becoming unemployed during contractions. These results suggest that any stress-induced increases in drinking during bad economic times are more than offset by declines resulting from changes in economic factors such as lower incomes.

Using genetic algorithms to assess the impact of pricing activity timing

Current methods of assessing competitive structure from aggregate sales data, such as the estimating of marketing mix elasticities, offer insight into the magnitude and direction of competitive actions. However, this level of analysis offers no insight into the impact that the timing of such mix activity has on one’s market share. In this research, I illustrate a method of assessing market structure by utilizing the timing of competitive mix activity. Specifically, using a genetic algorithm approach, I estimate marketing mix timing rules for a consumer product market. This research illustrates the ability of these marketing mix timing rules to assess the impact that the timing of competitive mix activity has on one’s market share. In addition, I show that these marketing mix timing rules offer strategic insight that can complement existing methods of assessing competitors using aggregate sales data, such as elasticity analysis, thereby improving the brand manager’s understanding of the competitive structure of the market.