Sage Journals: Discover world-class research

Abstract

Introduction. Consumption of sugar-sweetened beverages (SSBs) contributes to weight gain, obesity, and diabetes. Soda tax has been proposed to reduce consumption of SSBs. What remains unclear is whether the soda tax has an effect on health and health care costs. We evaluated the cost-effectiveness of a 1-cent-per-ounce soda tax on obesity and diabetes in California. Methods. A microsimulation state-transition model was used to evaluate the cost-effectiveness of the soda tax. Health outcomes were measured in quality-adjusted life-years (QALYs). Health care costs were projected from 2015 to 2035. Results. In a simulated cohort of Californian adults, the soda tax policy prevented 2.28 million cases of overweight (95% confidence interval [CI] −0.06 to 6.63) and 0.49 million cases of obesity (95% CI −0.19 to 1.18). From the health care perspective, the incremental cost-effectiveness ratio of the soda tax was $124,839 dollars per QALY (95% CI −1,151,983 to 557,660). From the health care perspective, the soda tax policy was cost-effective 80% of the time in the probabilistic sensitivity analysis using a willingness-to-pay threshold of $100,000 per QALY. Conclusions. The 1-cent-per-ounce soda tax reduced the number of obesity cases, diabetes cases, and related complications. In addition, the soda tax policy implemented in California was cost-effective most of the time.

Highlights

Question: What remains unclear is whether the soda tax has an effect on health and health care costs.

Findings: The 1-cent-per-ounce soda tax reduced the number of obesity cases, diabetes, and related complications. In addition, the soda tax policy brought large amounts of revenue.

Meaning: This study provides additional evidence regarding the health care costs and cost-effectiveness related to the implementation of a soda tax.

Keywords

soda tax obesity diabetes cost effectiveness analysis microsimulation California

Introduction

Obesity is a major public health problem in the United States, and its prevalence increased from 30.5% to 41.9% from 2000 to 2020.¹ Medical costs for adults who have obesity are significantly higher than for adults with a healthy weight.¹ Obesity-related conditions include heart disease, stroke, diabetes, and certain types of cancer.¹ About 1 in 10 Americans have diabetes, and approximately 90% to 95% of them are type 2 diabetes.²

It has been well established that sugar-sweetened beverage (SSB) consumption increases the risk of weight gain, obesity, diabetes, and cardiovascular diseases (CVDs).^3,4 SSBs are beverages containing added caloric sweeteners such as brown sugar, corn sweetener, corn syrup, dextrose, fructose, glucose, high-fructose corn syrup, honey, lactose, malt syrup, maltose, molasses, raw sugar, and sucrose.⁵ They include soft drinks, fruit drinks, energy drinks, sweetened waters, and presweetened coffees and teas.⁶ SSBs contribute to more than half of added sugar in the American diet.⁷ On average, American youth consume 143 calories from SSBs, and US adults consume about 145 calories from SSBs on any given day.⁵ The average US household spends an estimated $850 annually on soft drinks, for a total of $65 billion alone in 2010.⁸

One proposed solution to reducing SSB consumption is a soda excise tax, which is implemented as a fee per beverage unit (e.g., 1-cent-per-ounce) and imposed on the manufacturers, in contrast to sales taxes that are collected from the consumer at purchase.^9–13 Berkeley, California, was the first US jurisdiction to pass a 1-cent-per-ounce soda tax in 2014.¹⁴ Albany, Oakland, and San Francisco in California approved a similar 1-cent-per-ounce tax in 2016.^15,16 Evidence in these cities showed that the soda tax reduces SSB purchases as well as intake.^17,18

However, it remains unclear whether the soda tax has an effect on health and health care costs.¹⁹ There has been debate on whether the soda tax should be implemented at the state or federal level from beverage companies. Several states have passed legislation that preempts new local soda taxes.²⁰ California’s preemption law, which went into effect on June 28, 2018, banned more local soda taxes.²⁰

The objective of this study is to assess the cost-effectiveness of a 1-cent-per-ounce soda excise tax on preventing obesity and diabetes in California, where health outcomes were evaluated in 2 ways: 1) cases of health conditions prevented and 2) quality-adjusted life-years (QALYs). The expected total revenue from the tax was also calculated to inform the ongoing policy debate.²¹

Methods

We implemented a microsimulation state-transition model to evaluate the cost-effectiveness of the soda tax in California. We followed the best practices outlined in the Second Panel on Cost-Effectiveness in Health and Medicine.²² We followed the consolidated Health Economic Evaluation Reporting Standards (CHEERS) statement for the reporting of health economic evaluations²³ (Supplement Table 1). The model had 2 arms: a soda tax arm and a “do nothing” arm. The model took a health care system and government perspective.

Target Population and Setting

Our target population was Californian adults who were 18 y or older in 2015, the first year in which the soda tax took effect.²⁴ We used data from the California Health Interview Survey (CHIS), which is an annual cross-sectional survey of California households.¹ The CHIS sample is weighted to be representative of the noninstitutionalized population living in households in California.²⁵ A detailed description of the sampling and weighting process is available in the original report of the survey.²⁵ Body mass index (BMI; kg/m²) was calculated from self-reported height and weight and used to assess overweight (25 ≤ BMI < 30) and obesity (30 ≤ BMI) status.¹

There were 29,247,121 adults in California in 2015.²⁶ There were 20,511 adults surveyed in CHIS 2015. We ran a model based on 20,000 individual hypothetical people and rescaled the results to the actual number of adults in California in 2015 by multiplying the model estimates by 1,462.36, which was equal to 29,247,121 divided by 20,000. To assign baseline weight and height to simulated individuals, we randomly sampled from the joint probability distributions of weight and height to preserve the correlation and heterogeneity between individuals at baseline.

Soda Tax Arm and Base Case

We assumed the soda tax was applied in retail business entities including restaurants and vending machines. We evaluated the impact of the soda tax (compared with no soda tax) in the development of overweight, obesity, diabetes, related complications, and death (Figure 1).

Figure 1

Markov model for cost-effectiveness analysis of a one-cent-per-ounce soda tax. The soda tax affected soda consumption, which affected added sugar intake, calorie intake, and weight change. MI, myocardial infarction; ESRD, end stage renal diseases. The cycle length is 1 year and the time horizon of the model is 20 years.

To compute the weight change due to soda tax, we applied the estimate from a longitudinal household survey of beverage consumption in Oakland 1 y after the implementation of the 1-cent-per-ounce soda tax.²⁷ In the study, beverage consumption was estimated using the National Health and Nutrition Examination Survey (NHANES) Dietary Screener Questionnaire, then the total amount of added sugar from beverages was constructed from the National Cancer Institute algorithm.²⁷ Daily grams of added sugar from SSBs reduced 3.08 g (95% confidence interval [CI], −10.27, 4.11) among adults. There are roughly 4 calories in 1 g of sugar. Therefore, the amount of sugar converted to roughly 12.32 fewer calories per day (95% CI −41.08, 16.44). We estimated the weight loss with a static calorie-to-weight model in which every pound of body weight loss requires an energy deficit of about 3,500 calories.²⁸ Thus, a 12-calorie reduction in consumption was expected to result in a weight change of −0.56 kg (95% CI −1.93, 0.77) each year, which is a weight reduction of 0.56 kg. We assumed that soda consumption and weight would change in the first year of policy implementation. We also assumed that the impact of the soda tax remained consistent over the years.

In the meantime, adults in both arms naturally gained 0.45 to 0.91 kg each year²⁹ (Table 1). In the base case, there were no soda taxes.

Table 1

Model Input: Weight Loss, Tax Revenue, Cost, Utilities, and Transition Probabilities

Parameter	Value	Source
Natural weight gain, kg/y	0.45 to 0.91	Hutfless, 2013²⁹
Weight change due to reduction of soda consumption, kg/y	−0.56 (95% CI −1.93 to 0.77)	Cawley, 2020²⁷
Tax revenue from taxed beverages per year	$862,722,888 (±25% 647,042,166 to 1,078,403,610)	Cerrigione, 2022³⁰
Soda tax government administration cost per year	$8,627,229 (±25% 6,470,422 to 10,784,036)	Cerrigione, 2022³⁰
Health cost per person per year
Normal weight	0	assumption
Overweight	$1,320 (95% CI 1,069 to 1,607)	Ward, 2021³¹
Obesity	$3,955 (95% CI 3,519 to 4,363)	Ward, 2021³¹
Diabetes	$21,218 (±25% 15,913 to 26,522)	American Diabetes Association, 2018³²
Stroke	$103,425 (±25% 77,569 to 129,281)	Ward, 2014³³
Poststroke	$38,159 (±25% 28,619 to 47,698)	Ward, 2014³³
MI	$138,603 (±25% 103,952 to 173,253)	Ward, 2014³³
Post-MI	$4,675 (±25% 3,506 to 5,843)	Ward, 2014³³
ESRD	$176,647 (±25% 132,485 to 220,808)	National Institute of Diabetes and Digestive and Kidney Diseases, 2019³⁴
Death	0
Utilities
Normal weight	1 (±25% 0.75 to 1)	Assumption
Overweight	0.876 (±25% 0.657 to 1)	Jia, 2009³⁵
Obesity	0.850 (±25% 0.637 to 1)	Jia, 2009³⁵
Diabetes	0.785 (95% CI 0.681 to 0.889)	Beaudet, 2014³⁶
Stroke, poststroke	0.621 (95% CI 0.459 to 0.784)	Beaudet, 2014³⁶
MI, post-MI	0.730 (95% CI 0.614 to 0.847)	Beaudet, 2014³⁶
ESRD	0.61 (95% CI 0.51 to 0.71)	McManus, 2021³⁷
Death	0
Transition probabilities
Normal weight → diabetes	0.0069 (95% CI 0.0058 to 0.0083)	American Heart Association, 2022³⁸
Normal weight → stroke	0.0028 (95% CI 0.0025 to 0.0032)	American Heart Association, 2022³⁸
Normal weight → MI	0.0028 (95% CI 0.0025 to 0.0033)	American Heart Association, 2022³⁸
Normal weight → ESRD	0.0004 (±25% 0.0003 to 0.0005)	American Heart Association, 2022³⁸
Normal weight → death	0.0090 (±25% 0.0067 to 0.0112)	Centers for Disease Control and Prevention, 2019³⁹
Overweight → diabetes	0.0088 (95% CI 0.0084 to 0.0092)	Ganz, 2014⁴⁰
Overweight → stroke	0.0029 (95% CI 0.0027 to 0.0033)	Mitchell, 2015⁴¹
Overweight → MI	0.0033 (95% CI 0.0028 to 0.0039)	Thomsen, 2014⁴²
Overweight → ESRD	0.0005 (95% CI 0.0005 to 0.0006)	Wang, 2008⁴³
Overweight → death	0.0087 (95% CI 0.0087 to 0.0087)	Angelantonio, 2016⁴⁴
Obesity → diabetes	0.0135 (95% CI 0.0119 to 0.0154)	Cameron, 2021⁴⁵
Obesity → stroke	0.0035 (95% CI 0.0032 to 0.0039)	Mitchell, 2015⁴¹
Obesity → MI	0.0044 (95% CI 0.0034 to 0.0055)	Thomsen, 2014⁴²
Obesity → ESRD	0.0007 (95% CI 0.0006 to 0.0008)	Wang, 2008⁴³
Obesity → death	0.0118 (95% CI 0.0117 to 0.0119)	Angelantonio, 2016⁴⁴
Diabetes → stroke	0.0042 (95% CI 0.0032 to 0.0057)	Gregg, 2014⁴⁶
Diabetes → MI	0.0051 (95% CI 0.0038 to 0.0067)	Gregg, 2014⁴⁶
Diabetes → ESRD	0.0023 (95% CI 0.0022 to 0.0024)	Gregg, 2014⁴⁶
Diabetes → death	0.0128 (95% CI 0.0120 to 0.0137)	Li, 2019⁴⁷
Stroke → MI	0.03 (95% CI 0.01 to 0.05)	Gunnoo, 2015⁴⁸
Stroke → death	0.17 (±25% 0.13 to 0.21)	American Heart Association, 2022³⁸
MI → stroke	0.081 (±25% 0.061 to 0.101)	Loh, 1997⁴⁹
MI → death	0.22 (±25% 0.16 to 0.27)	American Heart Association, 2022³⁸
ESRD → stroke	0.0172 (95% CI 0.0146 to 0.0200)	Seliger, 2003⁵⁰
ESRD → MI	0.0040 (95% CI 0.0032 to 0.0050)	Beddhu, 2002⁵¹
ESRD → death	0.034 (±25% 0.025 to0.042)	Centers for Disease Control and Prevention, 2019³⁹
Poststroke → MI	0.0167 (95% CI 0.0136 to 0.0198)	Boulanger, 2018⁵²
Poststroke → death	0.0173 (95% CI 0.0119 to 0.0235)	Engstad, 2003⁵³
Post-MI → stroke	0.0087 (95% CI 0.0083 to 0.0092)	Sundboll, 2016⁵⁴
Post-MI → death	0.21 (±25% 0.16 to 0.26)	Law, 2002⁵⁵

±25%, plus and minus 25% of the value; CI, confidence interval; ESRD, end-stage renal disease; MI, myocardial infarction. → indicates the transition from one heath state to another health state. All costs were calculated in 2022 dollars.

Microsimulation State-Transition Model

Our microsimulation Markov model simulated the individual trajectories, conditional on previous history of disease.^56,57 The model had a 1-y cycle length and 20-y time horizon (2015–2035).

Health states included normal weight, overweight, obesity, diabetes, and death. We also included stroke, poststroke, myocardial infarction (MI), post-MI, and end-stage renal disease (ESRD) as diabetes complications. We differentiated first-time stroke from recurrent stroke and first-time MI from recurrent MI because first-time occurrence is more costly.³³

We followed the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework from the Cochrane Collaboration to develop summaries of model input parameters.^58,59 A summary of inputs and data sources of key model parameters is provided in Table 1. A detailed description of the search strategy can be found in Supplementary Text 1.

Transition Probabilities

Transition probabilities were obtained from previously published meta-analyses or prospective cohort studies. Following conversion formula denoted in Gidwani and Russell⁶⁰ and VanderWeele,⁶¹ we derived 1-y transition probabilities from sex-/age-/race-adjusted risks, risk ratios, incidence rate ratios, hazard ratios, or odds ratios (Supplementary Text 2, Supplementary Figure 1).

Cost and Utility Data

Cost data

Soda tax implementation costs included the government administration cost. MUNI Services, a private firm contracted by the city of Berkeley to collect soda tax revenue, charges 2% of the tax proceeds.²¹ Therefore, we estimated the overall cost of implementing the statewide soda tax policy as 2% of the annual soda tax revenue, with 1% for government administration cost.⁵⁰ Annual state revenue was calculated using the UConn Rudd Center for Food Policy and Health Sugary Drink Tax Calculator. This calculator multiplies the average per-person intake of SSB after soda tax implementation by the tax rate and the total adult population.³⁰ Beverage sales were assumed to represent beverage consumption. Proprietary data from the Beverage Marketing Corporation was used to measure the total sales of SSB sold across all retail channels in the United States, including all types of food stores, fountain drinks in restaurants, vending machines, and so forth.³⁰

Direct medical costs for overweight and obesity included medical expenditures for hospital stays, emergency room (ER) visits, outpatient department visits, office-based medical provider visits, home health care, other medical expenses, and prescription medicines.^31,62 Direct medical costs for diabetes included institutional care, outpatient care, oral medications, noninsulin injectable antidiabetes agents, hearing devices, eyewear, and prostheses.³² Direct medical costs for stroke, MI, and ESRD included inpatient stays, ER visits, outpatient visits, tests and procedures, as well as medications.³³ Direct medical costs were estimated from the United States Renal Data System 2022 Annual Data Report and the American Diabetes Association 2017 Report^32–34 (Table 1).

Utility data

We used EQ-5D utility to value health.⁶³ Normal weight was assigned a utility value of 1 and death a utility value of 0.⁵¹ The EQ-5D scores for other conditions were 0.876 for overweight, 0.850 for obesity, 0.785 (95% CI 0.681, 0.889) for diabetes, 0.621 (95% CI 0.459, 0.784) for stroke as well as poststroke, 0.730 (95% CI 0.614, 0.847) for MI as well as post-MI, and 0.61 (95% CI 0.51, 0.71) for ESRD.^35–37

All costs were adjusted for inflation to 2022 dollars using the Personal Consumption Expenditure Price Index.⁵⁷ Both costs and health outcomes were discounted at 3% per year.

Study Perspectives

Analyses were conducted from the health care sector as well as government perspective, similar to Lee et al.,^22,64 also in accordance with recommendations from the Second Panel on Cost-Effectiveness in Health and Medicine. For the health care sector perspective, we included only direct medical costs, and for the government perspective, we incorporated government tax collection costs, soda tax revenue, and health care costs.

Outcomes

Health gains included the number of new diseases prevented, deaths averted, and QALYs gained and cost-effectiveness was assessed using the incremental cost-effectiveness ratio (ICER). The ICER was calculated by dividing the difference in net costs by the difference in net effectiveness. The net costs and net effectiveness were computed comparing the costs and effectiveness (QALYs) in the soda tax scenario with the status quo (no soda tax).¹² We used a willingness-to-pay (WTP) threshold of $100,000 per QALY and presented the results graphically using cost-effectiveness acceptability curves.⁶⁵

Sensitivity Analysis

To evaluate the uncertainty around estimates and generate standard errors, we used probabilistic sensitivity analyses, using Monte Carlo sampling with replacement for all input parameters for 1,000 iterations (Table 1). We specified gamma distribution for cost inputs, log-normal distribution for relative risk, and beta distribution for probability and utility.^66,67 Results were presented as means and 95% CIs.

Internal and External Validation

We evaluated the internal validity of the model by comparing the weight distribution of the simulated cohort with the CHIS cohort in 2015. We also compared the weight distribution as well as the distribution of clinical outcomes of the simulated cohort with the CHIS cohort in 2019.

We evaluated the external validity of the model by changing the weight and height input to be the NHANES in 2015 and then compared the weight distribution as well as the distribution of clinical outcomes of the simulated cohort with the NHANES cohort in 2019.^68,69

Data analysis was implemented in R software, version 4.0.2.

Role of the Funding Source

The funder had no role in study design, data collection or analysis, decision to publish, or preparation of the article.

Results

Model Evaluation

The simulated distribution of weight broadly matched the observed distribution of weight from CHIS in 2015. Among 20,511 adults surveyed by CHIS in 2015, the mean weight was 77.67 kg (s = 19.96) and 5,794 (28.2%) were obese. In the simulated cohort in 2015, the mean weight was 77.97 kg (18.92) and 7,217 (25.9%) were obese (Supplementary Table 2). The simulation model-generated values also broadly matched the observed distribution of weight and clinical outcomes from the CHIS 2019 and NHANES 2019 database (Supplementary Tables 3 and 4).

Main Scenario

There would be 1.27 million (95% CI 1.04 to 1.54) new cases of diabetes when there was a soda tax, compared with 1.64 million (95% CI 1.16 to 2.43) new cases of diabetes when there was no soda tax. In addition, 3.28 million (95% CI −0.06 to 6.63) cases of overweight, 0.49 million cases of obesity (95% CI −0.19 to 1.18), 0.19 million new cases of stroke (95% CI −0.14 to 0.52), 0.16 million new cases of MI (95% CI −0.17 to 0.48), and 0.71 cases of deaths (95% CI −0.41 to 1.83) would be prevented from soda tax (Table 2). The total QALYs were 0.39 billion (95% CI 0.39 to 0.40).

Table 2

Summary of Comparativeness under Status Quo and Soda Tax for a Hypothetical Cohort of Californian Adults in 20 y (in Millions)^a

Items	1-Cent-per Ounce Soda Tax	Status Quo	Number of Cases Prevented
Number of overweight cases	10.53 (9.30, 14.08)	13.81 (12.92, 14.35)	3.28 (−0.06, 6.63)
Number of obesity cases	10.61 (10.53, 11.18)	11.10 (10.71, 11.71)	0.49 (−0.19, 1.18)
Number of diabetes cases	1.27 (1.04, 1.54)	1.64 (1.16, 2.43)	0.38 (−0.33, 1.08)
Number of stroke cases	0.45 (0.36, 0.54)	0.64 (0.42, 1.02)	0.19 (−0.14, 0.52)
Number of MI cases	0.83 (0.69, 0.99)	0.99 (0.75, 1.31)	0.16 (−0.17, 0.48)
Number of ESRD cases	0.30 (0.26, 0.34)	0.29 (0.25, 0.33)	−0.01 (−0.07, 0.05)
Number of death cases	2.03 (1.83, 2.23)	2.74 (1.97, 4.00)	0.71 (−0.41, 1.83)

Mean and 95% confidence interval. ESRD, end-stage renal disease; MI, myocardial infarction.

From a health care perspective, over 20 y, the health care cost was $529.70 billion (95% CI 420.28 to 681.19) in California when there was a soda tax. Health costs when there was no soda tax were $677.09 billion (95% CI 603.52 to 749.83). The total QALYs were 0.39 billion (95% CI 0.39 to 0.39). The resulting ICER was −$124,839 (95% CI −1,151,983 to 557,660) dollars per QALY (Table 3).

Table 3

Summary of Cost-Effectiveness under Status Quo and Soda Tax for a Hypothetical Cohort of Californian Adults in 20 y (Cost and QALYs in Billions)

Items	Status Quo	1-Cent-per Ounce Soda Tax	Incremental Changes
Total QALYs	0.39 (0.39, 0.39)	0.39 (0.39, 0.40)	—
Total health care costs	$677.09 (603.52, 749.83)	$529.70 (420.28, 681.19)	—
Health care cost savings per person	—	—	$ 0.01
Policy implementation cost	—	$ 0.17 (0.13, 0.22)	—
Total tax revenue	—	$18,609.77 (18,719.19, 18,458.28)	—
Net costs in health care perspective	—	—	−$147.39
Net costs in government perspective	—	—	−$19,286.69
ICER, health care perspective	—	—	−124,839 (−1,151,983 to 557,660)

ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life-year.

Data are presented as mean and 95% confidence interval. Negative costs represent net revenues. Costs are shown in 2022 dollars.

The tax revenue from SSBs was $862.72 million per year. Over 20 y, the net revenue for government was $18,609.60 billion (95% CI 18,719.02 to 18,458.11). Soda tax was cost saving for the government, and the ICER was −14,668,299 (95% CI −112,205,426 to −5,7206,440) dollars per QALY (Table 3).

Based on a WTP of $100,000/QALY, the soda tax policy was cost-effective 80% of the time (Figures 2 and 3).

Figure 2

Incremental cost-effectiveness (CE) plane for the 1-cent-per-ounce soda tax on obesity and overweight prevalence. The willingness-to-pay (WTP) threshold was defined as $100,000 per quality-adjusted life-year (QALY), health care perspective.

Figure 3

Cost-effectiveness acceptability curve (CEAC) for a 1-cent-per-ounce soda tax, health care perspective. Willingness to pay (WTP) is expressed in 2022 US dollars per quality-adjusted life-year gained ($/QALY). CEAC represents the probability for the 1-cent-per-ounce soda tax to be cost-effective (CE) option for different WTP thresholds.

Discussion

In this large simulation model of adults in California, we found that a soda tax reduced the number of obesity cases, diabetes, and related CVDs. Second, using the currently known intervention effect, we also found that the soda tax policy implemented in California was cost-effective in reducing the proportion of SSB-related overweight or obesity about 80% of the time from the health care perspective. Lastly, we found that the soda tax policy brought to the government large amounts of revenue totaling $18,609.77 billion over 20 y.

Both feasibility and effectiveness in reducing soda intake are supported by the 7 US cities and locales that have adopted soda tax. Seattle enacted a 1.75-cent-per-ounce soda tax in June 2017. The volume sold of taxed beverages fell in the first year following the tax in Seattle.⁷⁰ Results from Philadelphia are also promising, where the odds of daily consumption of regular soda were lower within the first 2 mo of tax implementation.⁷¹

Our study adds to the evidence base that instituting a soda tax has the potential of having a significant positive population impact on health and well-being by reducing cases of obesity, diabetes, and consequent complications.^{13,21,64,72,73} Ten-year-long nationwide simulations showed that a 1-cent-per-ounce soda tax reduced health care costs due to obesity.⁷⁴ The 1-cent-per-ounce soda tax was also expected to prevent a significant number of diabetes cases over the lifetime.⁶⁴ Such health benefits seemed to be greater when the amount of soda tax increased. A 2-cent-per-ounce soda tax was expected to prevent more obesity cases than 1-cent-per-ounce soda tax.^75,76 Therefore, it was possible to raise the amount of soda tax to further influence SSB consumption, thereby reducing calorie intake and weight.

Compared with other policies (e.g., the elimination of soda from vending machines or having health-related warning labels on soda packaging), the soda tax has the advantage of also generating significant revenues.⁷⁷ In our study, the proposed 1-cent-per-ounce soda tax was estimated to generate $862.72 million per year in California alone. Estimates from other data sources showed a similar large amount of revenues.^13,78 These revenues could be reinvested in preventive health activities. For instance, in Berkeley, California, revenue from soda tax has been allocated to school and community programs to promote healthy eating, increasing public support for such taxes.²⁴ Such reinvestment in prevention activities has the potential to produce larger health benefits. This revenue characteristic makes it a highly desirable option for policy makers weighing the costs and benefits of alternative approaches for addressing the urgent public health issue of obesity.

Basu et al.⁷⁹ and Long et al.¹³ illustrated that a soda tax is more cost-effective than soda bans and restrictions. Studies have also shown that a soda tax resulted in more health benefits and saved more on health care costs compared with a fruit and vegetable subsidy, reduced tax subsidy on TV advertising, and active physical education policies.^76,80 Societal cost savings were greater than $100,000 per 10,000 residents over 10 y when other health conditions—specifically coronary heart disease, cerebrovascular accident, chronic kidney disease, and dental caries and periodontal disease—were also included.⁷²

Strengths and Limitations

Our study has several strengths. We used state-representative data inputs. We assessed 20-y health effects, costs, and cost-effectiveness from health care as well as government perspectives. We modeled the impact of uncertainty in probabilistic sensitivity analyses.

Our project has some limitations. First, in the current study, we did not assess the soda tax’s cost-effectiveness from a societal perspective because soda tax revenues, which are of interest here, are costs that transferred from the public to the government and are not recommended to be included in the societal perspective.⁷⁴ Viable indirect cost estimates for conditions including MI, stroke, obesity, and overweight in the current literature were also not available. Second, there is strong evidence that changes in soda intake were also directly related to diabetes, stroke, and MI independent of weight change. Other mechanisms, such as hyperglycemia, dyslipidemia, inflammation, or endothelial dysfunction, underlie the association.^81–83 However, we focused on the effects of soda consumption on health outcomes through changes in weight. Therefore, our findings could represent an underestimation of health benefits and cost savings. Third, our results were based on a simulation model that incorporated a broad range of data inputs. While we included the best available evidence on population characteristics, trajectories of obesity prevalence, and obesity-related health care costs, our ability to forecast the precise impacts of soda tax was limited by the uncertainty around each of these inputs and by the assumptions required to build the model. We conducted probabilistic sensitivity analysis to evaluate the extent to which model input uncertainties affect the results. Fourth, we assumed that the effect of soda tax was the same across baseline characteristics (e.g., weight and gender). This was unrealistic because people from different backgrounds could be affected differently and, for example, adjust soda consumption based on their health status. Respondents in the study where we derived estimate of reductions of SSB consumption following soda tax were mostly middle-aged females, who consumed fewer SSB than younger populations did.^27,84 Therefore, our results could represent an underestimation of the impact of a 1-cent-per-ounce soda tax on weight.

Conclusion

The 1-cent-per-ounce soda tax reduced the number of obesity cases, diabetes, and related CVDs. In addition, the soda tax policy implemented in California was cost-effective most of the time from the health care perspective. This study provides additional evidence regarding the health care costs and cost-effectiveness related to the implementation of a soda tax.

Supplemental Material

sj-docx-1-mpp-10.1177_23814683241309669 – Supplemental material for Evaluation of the Soda Tax on Obesity and Diabetes in California: A Cost-Effectiveness Analysis

Supplemental material, sj-docx-1-mpp-10.1177_23814683241309669 for Evaluation of the Soda Tax on Obesity and Diabetes in California: A Cost-Effectiveness Analysis by Fan Zhao, Risha Gidwani, May C. Wang, Liwei Chen and Roch A. Nianogo in MDM Policy & Practice

Footnotes

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Financial support for this study was provided entirely by a grant from the National Institute on Minority Health and Health Disparities, National Institutes of Health, Bethesda, Maryland. The funding agreement ensured the authors’ independence in designing the study, interpreting the data, writing, and publishing the report.

Research Ethics

This study involves deidentified survey data and is not considered human subject research. Institutional review board approval was not needed.

Data Availability

The analysis codes for implementing the microsimulation model are available on GitHub.

ORCID iDs

Fan Zhao

Risha Gidwani

References

Centers for Disease Control and Prevention. Obesity is a common, serious, and costly disease. February 2021. Available from: https://www.cdc.gov/obesity/data/adult.html. [Accessed 19 April, 2021].

Centers for Disease Control and Prevention. Type 2 diabetes. May 2019. Available from: https://www.cdc.gov/diabetes/basics/type2.html. [Accessed 19 April, 2021].

de Koning

Malik

Kellogg

Rimm

Willett

FB.

Sweetened beverage consumption, incident coronary heart disease, and biomarkers of risk in men. Circulation. 2012;125(14):1735–41.

Malik

Popkin

Bray

Després

FB.

Sugar-sweetened beverages, obesity, type 2 diabetes mellitus, and cardiovascular disease risk. Circulation. 2010;121(11):1356–64.

Centers for Disease Control and Prevention. Sugar sweetened beverage intake. March 2021. Available from: https://www.cdc.gov/nutrition/data-statistics/sugar-sweetened-beverages-intake.html. [Accessed 3 April, 2021].

Department of Health. Sugar sweetened beverages. Available from: https://health.ri.gov/healthrisks/sugarsweetenedbeverages/. [Accessed 3 April, 2021].

Bailey

Fulgoni

Cowan

Gaine

PC.

Sources of added sugars in young children, adolescents, and adults with low and high intakes of added sugars. Nutrients. 2018;10(1):102.

Tuttle

How much you spend each year on coffee, gas, Christmas, pets, beer, and more. Time. January 23, 2012. Available from: https://business.time.com/2012/01/23/how-much-you-spend-each-year-on-coffee-gas-christmas-pets-beer-and-more/. [Accessed 30 July, 2021].

Berkey

Rockett

HRH

Field

Gillman

Colditz

GA.

Sugar-added beverages and adolescent weight change. Obes Res. 2004;12(5):778–88.

10.

Malik

VS.

Sugar-sweetened beverages and risk of obesity and type 2 diabetes: epidemiologic evidence. Physiol Behav. 2010;100(1):47–54.

11.

Schulze

Manson

Ludwig

, et al. Sugar-sweetened beverages, weight gain, and incidence of type 2 diabetes in young and middle-aged women. JAMA. 2004;292(8):927–34.

12.

Chen

Appel

Loria

, et al. Reduction in consumption of sugar-sweetened beverages is associated with weight loss: the PREMIER trial. Am J Clin Nutr. 2009;89(5):1299–306.

13.

Long

Gortmaker

Ward

, et al. Cost effectiveness of a sugar-sweetened beverage excise tax in the U.S. Am J Prev Med. 2015;49(1):112–23.

14.

Urban Institute. Soda taxes. Available from: https://www.urban.org/policy-centers/cross-center-initiatives/state-and-local-finance-initiative/state-and-local-backgrounders/soda-taxes. [Accessed July 15, 2022].

15.

City of Albany, CA. Sugar sweetened beverage tax. Available from: https://www.albanyca.org/departments/finance/sugar-sweetened-beverage-tax. [Accessed April 12, 2021].

16.

City of Oakland, CA. Sugar sweetened beverage tax. Available from: https://bouldercolorado.gov/tax-license/finance-sugar-sweetened-beverage-tax. [Accessed April 28, 2021].

17.

Falbe

Thompson

Becker

Rojas

McCulloch

Madsen

KA.

Impact of the Berkeley excise tax on sugar-sweetened beverage consumption. Am J Public Health. 2016;106(10):1865–71.

18.

Silver

Ryan-Ibarra

, et al. Changes in prices, sales, consumer spending, and beverage consumption one year after a tax on sugar-sweetened beverages in Berkeley, California, US: a before-and-after study. PLoS Med. 2017;14(4):e1002283.

19.

Hattersley

Thiebaud

Silver

Mandeville

. Countering common arguments against Taxes on sugary drinks. June 2020. Available from: http://hdl.handle.net/10986/34361. [Accessed May 1, 2023]

20.

Crosbie

Pomeranz

Wright

Hoeper

Schmidt

State preemption: an emerging threat to local sugar-sweetened beverage taxation. Am J Public Health. 2021;111(4):677–86.

21.

Wilde

Huang

, et al. Cost-effectiveness of a US national sugar-sweetened beverage tax with a multistakeholder approach: who pays and who benefits. Am J Public Health. 2019;109(2):276–84.

22.

Sanders

Neumann

Basu

, et al. Recommendations for conduct, methodological practices, and reporting of cost-effectiveness analyses: second panel on cost-effectiveness in health and medicine. JAMA. 2016;316(10):1093–103.

23.

Husereau

Drummond

Petrou

, et al. Consolidated Health Economic Evaluation Reporting Standards (CHEERS)—explanation and elaboration: a report of the ISPOR health economic evaluation publication guidelines good reporting practices task force. Value Health. 2013;16(2):231–50.

24.

City of Berkeley. Berkeley’s tax ordinance. January 2015. Available from: http://www.healthyberkeley.com/about-berkeleys-tax-ordinance. [Accessed April 12, 2021].

25.

UCLA Center for Health Policy Research. Get CHIS data. Available from: https://healthpolicy.ucla.edu/chis/data/Pages/confidential.aspx. [Accessed April 30, 2021].

26.

United States Census Bureau. US Census. Available from: https://data.census.gov/cedsci/table?q=B03002&g=0400000US06&tid=ACSDT1Y2018.B03002&hidePreview=true. [Accessed April 3, 2021].

27.

Cawley

Frisvold

Hill

Jones

Oakland’s sugar-sweetened beverage tax: impacts on prices, purchases and consumption by adults and children. Econ Hum Biol. 2020;37:100865.

28.

Hall

Sacks

Chandramohan

, et al. Quantification of the effect of energy imbalance on bodyweight. Lancet. 2011;378(9793):826–37.

29.

Hutfless

Maruthur

Wilson

, et al. Strategies to prevent weight gain among adult. 2013. Available from: https://www.ncbi.nlm.nih.gov/books/NBK133218. [Accessed May 19, 2023].

30.

Cerrigione C. Tax calculator. UConn Rudd Center for Food Policy & Obesity. 2020. Available from: https://uconnruddcenter.org/tax-calculator/. [Accessed August 22, 2021].

31.

Ward

Bleich

Long

Gortmaker

SL.

Association of body mass index with health care expenditures in the United States by age and sex. PLoS One. 2021;16(3):e0247307.

32.

American Diabetes Association. Economic costs of diabetes in the U.S. in 2017. Diabetes Care. 2018;41(5):917–28.

33.

Ward

Alvarez

Martin

Direct medical costs of complications of diabetes in the United States: estimates for event-year and annual state costs (USD 2012). J Med Econ. 2014;17(3):176–83.

34.

National Institute of Diabetes and Digestive and Kidney Diseases. Kidney Disease Statistics for the United States. Available from: https://www.niddk.nih.gov/health-information/health-statistics/kidney-disease. [Accessed January 20, 2023].

35.

Jia

Lubetkin

EI.

The statewide burden of obesity, smoking, low income and chronic diseases in the United States. J Public Health. 2009;31(4):496–505.

36.

Beaudet

Clegg

Thuresson

Lloyd

McEwan

Review of utility values for economic modeling in type 2 diabetes. Value Health. 2014;17(4):462–70.

37.

McManus

Lee

Hur

Kuo

JH.

Timing of parathyroidectomy for tertiary hyperparathyroidism with end-stage renal disease: a cost-effectiveness analysis. Surgery. 2021;169(1):94–101.

38.

Tsao

Aday

Almarzooq

, et al. Heart disease and stroke statistics—2022 update: a report from the American Heart Association. Circulation. 2022;145(8):e153–639.

39.

Centers for Disease Control and Prevention. Compressed Mortality, 1999-2016 Request. 2019. Available from: https://wonder.cdc.gov/cmf-icd10.html. [Accessed 16 May, 2023].

40.

Ganz

Wintfeld

Alas

Langer

Hammer

The association of body mass index with the risk of type 2 diabetes: a case–control study nested in an electronic health records system in the United States. Diabetol Metab Syndr. 2014;6(1):50.

41.

Mitchell

Cole

McArdle

, et al. Obesity increases risk of ischemic stroke in young adults. Stroke. 2015;46(6):1690–2.

42.

Thomsen

Nordestgaard

BG.

Myocardial infarction and ischemic heart disease in overweight and obesity with and without metabolic syndrome. JAMA Intern Med. 2014;174(1):15–22.

43.

Wang

Chen

Song

Caballero

Cheskin

LJ.

Association between obesity and kidney disease: a systematic review and meta-analysis. Kidney Int. 2008;73(1):19–33.

44.

Angelantonio

Bhupathiraju

Wormser

, et al. Body-mass index and all-cause mortality: individual-participant-data meta-analysis of 239 prospective studies in four continents. Lancet. 2016;388(10046):776–86.

45.

Cameron

Petito

McCabe

, et al. Quantifying the sex-race/ethnicity-specific burden of obesity on incident diabetes mellitus in the United States, 2001 to 2016: MESA and NHANES. J Am Heart Assoc. 2021;10(4):e018799.

46.

Gregg

Wang

, et al. Changes in diabetes-related complications in the United States, 1990–2010. N Engl J Med. 2014;370(16):1514–23.

47.

Wang

Zhang

Liu

Diabetes mellitus and cause-specific mortality: a population-based study. Diabetes Metab J. 2019;43(3):319–41.

48.

Gunnoo

Hasan

Khan

Slark

Bentley

Sharma

Quantifying the risk of heart disease following acute ischaemic stroke: a meta-analysis of over 50 000 participants. BMJ Open. 2016;6(1):e009535.

49.

Loh

Sutton

Wun

, et al. Ventricular dysfunction and the risk of stroke after myocardial infarction. N Engl J Med. 1997;336(4):251–7.

50.

Seliger

Gillen

Longstreth

Kestenbaum

Stehman-Breen

CO.

Elevated risk of stroke among patients with end-stage renal disease. Kidney Int. 2003;64(2):603–9.

51.

Beddhu

Allen-Brady

Cheung

, et al. Impact of renal failure on the risk of myocardial infarction and death. Kidney Int. 2002;62(5):1776–83.

52.

Boulanger

Béjot

Rothwell

Touzé

Long-term risk of myocardial infarction compared to recurrent stroke after transient ischemic attack and ischemic stroke: systematic review and meta-analysis. J Am Heart Assoc. 2018;7(2):e007267.

53.

Engstad

Viitanen

Arnesen

Predictors of death among long-term stroke survivors. Stroke. 2003;34(12):2876–80.

54.

Sundbøll

Horváth-Puhó

Schmidt

, et al. Long-term risk of stroke in myocardial infarction survivors. Stroke. 2016;47(7):1727–33.

55.

Law

Watt

Wald

NJ.

The underlying risk of death after myocardial infarction in the absence of treatment. Arch Intern Med. 2002;162(21):2405–10.

56.

Weinstein

Coxson

Williams

Pass

Stason

Goldman

. Forecasting coronary heart disease incidence, mortality, and cost: the Coronary Heart Disease Policy Model. Am J Public Health. 1987;77(11):1417–26.

57.

US Department of Veterans Affairs. Decision analysis: decision trees, simulation models, sensitivity analyses. Available from: https://www.herc.research.va.gov/include/page.asp?id=decision-analysis#simulation-models. [Accessed April 22, 2021].

58.

Guyatt

Oxman

Kunz

Vist

Falck-Ytter

Schünemann

HJ.

What is “quality of evidence” and why is it important to clinicians?

BMJ. 2008;336(7651):995–8.

59.

Higgins

Thomas

Chandler

, et al. Cochrane handbook for systematic reviews of interventions. Available from: https://training.cochrane.org/handbook/current. [Accessed April 20, 2021].

60.

Gidwani

Russell

LB.

Estimating transition probabilities from published evidence: a tutorial for decision modelers. Pharmacoeconomics. 2020;38(11):1153–64.

61.

VanderWeele

TJ.

Optimal approximate conversions of odds ratios and hazard ratios to risk ratios. Biometrics. 2020;76(3):746–52.

62.

Agency for Healthcare Research and Quality. Medical expenditure panel survey topics. Available from: https://meps.ahrq.gov/mepsweb/data_stats/MEPS_topics.jsp?topicid=33Z-1. [Accessed May 25, 2023].

63.

Muennig

Bounthavong

Cost-Effectiveness Analysis in Health: A Practical Approach. 3rd ed. New York: Wiley; 2016. Available from: https://www.wiley.com/en-us/Cost+Effectiveness+Analysis+in+Health%3A+A+Practical+Approach%2C+3rd+Edition-p-9781119011262. [Accessed 2023 Mar 17].

64.

Lee

Mozaffarian

, et al. Health impact and cost-effectiveness of volume, tiered, and absolute sugar content sugar-sweetened beverage tax policies in the United States. Circulation. 2020;142(6):523–34.

65.

Menzel

PT.

How should willingness-to-pay values of quality-adjusted life-years be updated and according to whom?

AMA J Ethics. 2021;23(8):601–6.

66.

Briggs

Weinstein

Fenwick

EAL

Karnon

Sculpher

Paltiel

AD.

Model parameter estimation and uncertainty analysis: a report of the ISPOR-SMDM modeling good research practices task force working group–6. Med Decis Making. 2012;32(5):722–32.

67.

Turner

Sanyal

Martin

Tannenbaum

Economic evaluation of sedative deprescribing in older adults by community pharmacists. J Gerontol A. 2021;76(6):1061–7.

68.

Health, United States, 2019. Centers for Disease Control and Prevention (U.S.). 2021. Available from: https://stacks.cdc.gov/view/cdc/100685. [Accessed July 21, 2024].

69.

Fryar

Kruszon-Moran

Ogden

CL.

Mean body weight, height, waist circumference, and body mass index among adults: United States, 1999–2000 through 2015–2016. Report No. 122. Atlanta (GA): Centers for Disease Control and Prevention; 2018.

70.

Powell

Leider

The impact of Seattle’s sweetened beverage tax on beverage prices and volume sold. Econ Hum Biol. 2020;37:100856.

71.

Zhong

Auchincloss

Lee

Kanter

GP.

The short-term impacts of the Philadelphia beverage tax on beverage consumption. Am J Prev Med. 2018;55(1):26–34.

72.

White

Basu

Kaplan

Madsen

Villas-Boas

Schillinger

Evaluation of the sugar-sweetened beverage tax in Oakland, United States, 2015–2019: a quasi-experimental and cost-effectiveness study. PLoS Med. 2023;20(4):e1004212.

73.

Wang

Coxson

Shen

Goldman

Bibbins-Domingo

A penny-per-ounce tax on sugar-sweetened beverages would cut health and cost burdens of diabetes. Health Aff (Millwood). 2012;31(1):199–207.

74.

Gortmaker

Long

Resch

, et al. Cost effectiveness of childhood obesity interventions: evidence and methods for CHOICES. Am J Prev Med. 2015;49(1):102–11.

75.

Gouck

Whetstone

Walter

, et al. Cost-effectiveness of a sugary drink excise tax in California. Available from: https://choicesproject.org/publications/report-california-sugary-drink-excise-tax/. [Accessed July 15, 2022].

76.

Long

Polacsek

Bruno

, et al. Cost-effectiveness analysis and stakeholder evaluation of 2 obesity prevention policies in Maine, US. J Nutr Educ Behav. 2019;51(10):1177–87.

77.

Kristensen

Flottemesch

Maciosek

, et al. Reducing childhood obesity through U.S. federal policy. Am J Prev Med. 2014;47(5):604–12.

78.

Andreyeva

Chaloupka

Brownell

KD.

Estimating the potential of taxes on sugar-sweetened beverages to reduce consumption and generate revenue. Prev Med. 2011;52(6):413–6.

79.

Basu

Seligman

Bhattacharya

. Nutritional policy changes in the supplemental nutrition assistance program: a microsimulation and cost-effectiveness analysis. Med Decis Making. 201333(7):937–48.

80.

Cobiac

Tam

Veerman

Blakely

Taxes and subsidies for improving diet and population health in Australia: a cost-effectiveness modelling study. PLoS Med. 2017;14(2):e1002232.

81.

Havard School of Public Health. Sugary drinks. September 2013. Available from: https://www.hsph.harvard.edu/nutritionsource/healthy-drinks/sugary-drinks/. [Accessed March 30, 2021].

82.

Torres-Ibarra

Rivera-Paredez

Hernández-López

, et al. Regular consumption of soft drinks is associated with type 2 diabetes incidence in Mexican adults: findings from a prospective cohort study. Nutr J. 2020;19(1):126.

83.

Gardener

Moon

Rundek

Elkind

MSV

Sacco

RL.

Diet soda and sugar-sweetened soda consumption in relation to incident diabetes in the northern Manhattan study. Curr Dev Nutr. 2018;2(5):nzy008.

84.

Rosinger

Herrick

Gahche

Park

. Sugar-sweetened beverage consumption among U.S. adults, 2011–2014. June 2019. Available from: https://www.cdc.gov/nchs/products/databriefs/db270.htm. [Accessed May 31, 2023].

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB