Food for Thought and Action: Responding to the Health Threats Posed by Climate Change

Health Letter, April 2021

By Michael T. Abrams, M.P.H., Ph.D.

In December 2020, the flagship health services research journal Health Affairs devoted its content to the human health effects of climate change. Notably per Alan Weil, the journal’s editor, this is the first time Health Affairs has published a full issue on the intersection of these current and vital issues.

Much of the issue’s content is a technical summary focused on methods and uncertainties that characterize this complex dual topic, the study of which requires the coupling of scientific disciplines from biology to economics and further must rely on data and projections spanning many years backwards and forwards in time. A good portion looks directly at the carbon footprint of the health care system (such as hospital operations and pharmaceutical production).

Overall, review of this material appropriately raises almost as many questions as observations, but some practical facts and concepts are presented throughout, and one article — about school lunch programs in the U.S. — was particularly useful in answering the question: “What might the average person do to address climate change?” The answer, to paraphrase Hippocrates, seems per that article to include the following: “Sometimes food is the best medicine.”

About climate change and health

An overview by Ebi and Hess states that the health risks related to human-caused climate change include heat waves, droughts, wildfires, storms, floods, infectious disease outbreaks (conveyed by contaminated food, water or insects) and general resource scarcities, all of which have become increasingly evident. These changes contribute to worldwide hunger, which has increased from 797 million affected people in 2016 to 821 million in 2018, and more than 2 billion people lack regular access to sufficient quantities of safe, nutritious food. Air pollution from emissions from power plants, industrial processes and agricultural activities, among others, causes approximately 4.5 million deaths globally each year. Adaptations are needed, but have been implemented using only 0.5% of the international financing currently addressing climate change.

U.S. health care sector greenhouse gas emissions rose 6% between 2010 to 2018 to 1,692 kg of CO2 per person in 2018, the highest among industrialized nations. In 2018 those emissions resulted in an estimated loss of approximately 388,000 disability-adjusted life-years. Production of pharmaceutical and other medical chemicals accounted for nearly 20% of those greenhouse gas emissions, the largest single emissions source identified. Adaptation in some large health-care systems has yielded improvements. Two of the most successful programs in that regard are the National Health Service in the U.K., which reduced its carbon emissions by 19% between 2007 and 2017, and the U.S. health management organization Kaiser Permanente, which yielded analogous reductions of 29% between 2008 and 2018.

Cost-benefit analysis is routinely used to judge the impact of regulatory decisions, including those directed at climate change mitigation. Such calculations require debatable economic assumptions about the distribution of health costs and benefits between rich and poor people and between current and future generations. Accordingly, Scovronick et al. ran seven distinct simulations to estimate a credible social cost range for CO2 emissions. The 2030 social cost thus was estimated to range from $22 to $139 per ton of CO2, with higher values connected to higher regard for poor people and future generations. The Trump administration calculated the social cost of CO2 at just $1-$7 per ton, markedly lower because they devalued future generations and people from other countries.

Natural disaster research affords acute and dramatic information regarding the human health costs of climate change. Hurricane Dorian struck the Bahamas in September 2019; the 185-mile-per-hour winds were the strongest seen there since 1935, and the storm pummeled the islands for two days, washing away homes, killing dozens and vanishing hundreds more. The storm’s aftermath subjected the islands to power and sanitation outages and water and food shortages. In the wake of this storm, presumably made especially intense and long-lasting because of global warming, volunteer medical personnel observed high rates of psychological trauma and the exacerbation of noncommunicable diseases, “most notably diabetes and hypertension.”

The Caribbean region was especially hard hit by two Category 5 storms, Hurricanes Irma and Maria, in 2017. These storms killed more than 3,000 people and caused $280 billion in property damage. Consistent with the Dorian experience in the Bahamas, a large proportion (30%) of the hurricane-related deaths seen in the wake of these two storms were due to complications from noncommunicable diseases such as diabetes, asthma, hypertension, cardiovascular disease and mental health disorders.

A 15-year study followed approximately 1,000 low-income Black residents (age 18-34 in 2004) who had experienced Hurricane Katrina, which struck New Orleans in August 2005. That study found that one week after Katrina, 45% of those already vulnerable because of a history of psychological distress could not obtain needed medical care, compared with 21% of those without such vulnerabilities (still a high proportion). Additionally, even access to medications was greatly curtailed, as 44% of those with psychological distress and 28% of all others could not obtain needed medications in the week after Katrina struck. Even four years later, 32% of all Black Hurricane Katrina survivors interviewed were still experiencing post-traumatic stress related to that storm, and 12 years later 17% were still affected. These are substantive indicators of the acute and longer-term negative health impacts of climate change with storms as the precipitating cause, and with mental health indicators as both important risk factors and consequences.

School lunch program study

As a specific incentive and guide towards reducing societal CO2 emissions in the U.S., one smart study analyzed existing menus served by the nation’s school lunch programs. The analysis involved comparisons of a composite of those actual menus to recommended standard menus designed by 37 experts in the nutritional, climate and agricultural sciences to maintain caloric intake and optimize nutrition and environmental sustainability. The standard is known as the EAT-Lancet reference diet.

The primary result of this work was a list of 18 food groups (dairy, dark green vegetables, fruit, whole grains, etc.), the amount allocated (in grams) and the calculated difference between the actual and recommended menus. As a secondary outcome, the total cost of the foods for the existing composite menus were calculated and compared to the age-appropriate EAT-Lancet menu costs.

Raw data used to generate these results came from a 2015 survey of school nutritional managers across 1,207 nationally representative elementary, middle and high schools. The survey respondents described five days of actual lunch menus for a specific week. Food prices were obtained from a U.S. Department of Agriculture database from 2003-2004 and subsequently adjusted for inflation to 2014. Separate calculations were done for each of these three educational levels to allow for broadly age-specific menus and serving sizes.

The table below summarizes the comparative intake results for middle schoolers. For example, the table shows that the actual intake of dairy in such schools was 232 grams more than the EAT-Lancet recommended amount and further that the absolute recommended dairy intake is 67 grams (and 16% of the total diet by weight). Accordingly, the center column records the average difference between the actual middle school and EAT-Lancet menus, a plus (+) sign indicating students are consuming more than recommended, and a minus (-) sign indicating the opposite. The last column of the table provides absolute amounts (in grams) and relative proportions (in percent) to guide one towards a healthier more environmentally sustainable diet.

Table. Actual U.S. Middle School* Lunch Composition Versus a Sustainable Benchmark (EAT-Lancet)

Food Category Actual Minus Reference Lunch (Grams)** EAT-Lancet (Recommended) Lunch (Grams) [Percent of Total]
Whole grains -69 110 [26%]
Dairy +232 67 [16%]
Fruit +94 53 [13%]
Legumes (beans, lentils, peas) -30 34 [8%]
Dark green vegetables -14 27 [6%]
Other vegetables -13 27 [6%]
Red, orange vegetables +8 27 [6%]
Peanuts, tree nuts -13 13 [3%]
Tubers, starchy vegetables +10 13 [3%]
Unsaturated oils -3 11 [3%]
Seafood -7 8 [2%]
Poultry (chicken and other) +8 8 [2%]
Added sugars +11 8 [2%]
Soy foods -4 7 [1%-2%]
Eggs -3 4 [1%]
Meats (beef, lamb, pork) +15 4 [1%]
Solid fats +3 3 [<1%]
Refined grains +26 0 [0%]
Total 424 {15 ounces} [100%]

*Based on 384 reporting middle schools.
**“+” shows that middle schoolers, on average, are served too much, “-” means they are not served enough within the food category. (28 grams ~ 1 ounce)
Source: Supplementary materials (Appendix Exhibit D) from Poole et al., Health Affairs, 2020;39:12:2144.

The overall patterns observed for elementary and high school are similar to those for middle school. Finally, pricing of the actual and recommended menus show that the EAT-Lancet diet is less expensive by 32% to 60%, thereby refuting the concern it will be more expensive than the status quo. Accordingly, the main challenge of implementing the school-based menu changes suggested by this study will likely be resistance from agricultural producers deeply invested in milk, meat and processed grain production. Changing consumption habits of students and preparation strategies of school personnel will also present as challenges.

Still, at least four reasons now exist to alter school diets towards the EAT-Lancet approach: 1) better individual nutrition, 2) financial savings on food acquisition, 3) reductions in aggregate CO2 emissions and 4) the broader societal health benefits that result from those reductions.

These reasons are not likely to shift our collective tastes from cheeseburgers to kale-and-tofu salads overnight, but the school lunch program is one logical place to motivate and educate 30 million students in the U.S. and others towards a more sustainable earth.