Evaluation of School Wellness Policies in Low-Income California Districts after the 2016 USDA Final Rule

Abstract

BACKGROUND:

Districts with federal nutrition programs must have an updated local school wellness policy (LSWP), to promote nutrition, physical activity, and student wellness. This study evaluates factors associated with LSWP quality among low-income districts.

METHODS:

In 2018, we collected LSWPs from websites of 200 randomly selected, county-stratified, low-income-serving California public districts. Multivariable linear regression assessed associations between district characteristics, model LSWP use (national, state, none), and adoption date on policy quality.

RESULTS:

On the WellSAT 3.0 scale of 0–100, mean (95% CI) comprehensiveness was 65.0 (63.2–66.7) and strength was 37.3 (35.3–39.2). Nearly verbatim, 68.5% adopted the state model LSWP, 13.0% adopted a national model, and half were adopted before mandated updates. District size (≥1,000 students) and national model LSWP adoption were associated with higher comprehensive scores. National model LSWP adoption was associated with higher strength scores in updated policies compared with those not updated.

IMPLICATIONS:

LSWPs have improved school food and activity environments, but district engagement in LSWP is low. Integration into education frameworks that reduce learning barriers could provide synergy for re-engagement.

CONCLUSIONS:

High adoption of model policies and low update compliance indicate little district engagement in LSWP. Mixed methods studies of districts with high quality LSWP are needed.

BACKGROUND

Poor diet quality and inactivity in childhood are associated with the development of obesity-related chronic diseases as well as physical and mental health challenges that can affect growth and development and impede academic success.^{1 2 3} Schools can play an important role in promoting child health by improving the quality of foods and beverages offered at school, increasing access to low-cost and free nutritious school meals, and supporting positive health behaviors like physical activity that promote health and learning.^{4 5}

Food and beverage offerings on school campuses include those served through the National School Lunch (NSLP) and School Breakfast Programs (SBP), those sold at school outside of these programs (competitive foods), and those available during school celebrations. Providing this support at school is crucial for low-income students who are at higher risk for both food insecurity and obesity-related disease. Low-income students are eligible to receive free and reduced-price meals (FRPM) through the NSLP and SBP. Students with family incomes less than 130% of the Federal Poverty Level are eligible for free meals, and those between 130 and 185% are eligible for reduced-priced meals.⁵ Low-income students make up about three-quarters of school meal participants and many consume their primary meal and more than half of their daily calories at school.^{4 5 6}

Recognizing this critical role, Congress passed the Child Nutrition and WIC Reauthorization Act in 2004, requiring all public-school districts that participate in federal school meal programs to develop and adopt a local school wellness policy (LSWP) by the 2006–2007 academic year.⁷ Each district policy must include (1) written goals for nutrition education, physical activity, and other wellness promotion activities; (2) nutrition guidelines for all foods served and sold on school campuses; (3) assurance that school meals meet federal requirements; (4) a plan for measuring implementation; and (5) involve parents, students, school food authority, school board, school administrators, and the public in the LSWP development.⁷ National and state school nutrition organizations offer resources including model policies to engage districts in LSWP development. As long as all required components are included, districts are encouraged to develop and customize their LSWP to meet local needs.⁸

In 2010, the Healthy, Hunger-Free Kids Act (HHFKA) updated and strengthened LSWPs by giving the U.S. Department of Agriculture (USDA) authority to set nutrition standards for all foods and beverages sold during the school day.⁹ The HHFKA also requires regulation of food and beverage marketing on campus, increased access to drinking water, standards for nutrition education and physical activity, and broader district wellness team membership, assessment and oversight. Districts are also required to review and revise LSWPs every three years, and make them available to the public.^{8 9} In July of 2016, the USDA released the final rule which codified HHFKA nutrition standards and mandated districts to revise their LSWPs to comply with the updated requirements HHFHby June 30, 2017.¹⁰ Multiple post HHFKA evaluations document improvements in the nutritional value of school meals and competitive food and beverage offerings as well as student purchasing of these items at school. ^{11 12 13 14}

While most districts have adopted a LSWP and policy quality has improved over time,¹⁵ LSWP assessments show on average, low ratings of policy comprehensiveness (number of policy components addressed) and strength (decisiveness of policy language).^{15 16 17} A recent analysis of a national sample using the validated, HHFKA-updated WellSat 3.0 LSWP scoring tool reported an average comprehensiveness score of 54/100 and strength score of 33/100,¹⁸ indicating on average, policies addressed about half of the components and used strong and decisive wording for only about a third.

Interest in improving LSWP quality persists because strong and comprehensive policy language indicates commitment and accountability and is associated with better policy implementation at the school level.¹⁹ To date, several studies found stronger policies associated with larger district size, majority Hispanic/Latino and Black districts, presence of district high schools, urban locale, and districts in states requiring additional LSWP mandates. ^{15 16 17} There are few evaluations of school wellness policies after the 2016 final rule update.

California (CA) is an excellent setting to study LSWP due to the size and diversity of the more than 6 million K-12 students. Nearly 3.6 million CA students qualify for the FRPM program, a key poverty indicator.²⁰ In addition, more than 60% represent racial/ethnic minorities and underserved demographic groups associated with the highest risk for both food insecurity and obesity-related disease.²¹ This is the first large scale LSWP evaluation in CA.

This study describes the distribution of LSWP comprehensiveness and strength scores across a county-stratified, randomly selected sample of CA public districts serving low-income students collected spring 2018. The goal is to examine whether district size, FRPM eligibility, a high percent of racial/ethnic minority students, urban locale, having a district high school, use of a model LSWP template, or policy adoption after passage of the 2016 final rule were associated with higher policy comprehensiveness and strength scores. The study focuses on districts serving low-income students who would benefit most from improvements tied to stronger LSWPs to inform future LSWP guidance and technical assistance efforts.

METHODS

Participants

Among California’s 1028 public districts, 770 met the “low-income” criteria, defined as having at least one school in the district with 50.0% or more FRPM-eligible students. These 770 districts were stratified by size and county to ensure adequate representation from each county in the final sample. Within each stratum 25% of districts were randomly selected with probability proportional to size, resulting in a final sample of 200 districts.

In spring 2018 the most recent LSWP was obtained from each of the 200 districts’ websites using established methods via internet search.¹⁵ Missing policies were requested from district administrators via phone and email. District demographic data were obtained from the California Department of Education (CDE).²² National Center for Education Statistics data were used to classify district urbanicity as urban or rural.^{23 24}

Instrumentation

The Rudd Center’s WellSAT 3.0, a validated, quantitative assessment tool, was used to score the policies because it captured both federal and state nutrition and physical activity requirements and has been used in other LSWP surveillance projects.^{18 25} WellSAT 3.0 includes 67 questions organized into six sections: (1) Nutrition Education; (2) Standards for USDA Child Nutrition Programs and School Meals; (3) Nutrition Standards for Competitive and Other Foods and Beverages; (4) Physical Education and Activity; (5) Wellness Promotion and Marketing; and (6) Implementation, Evaluation and Communication.²⁶ WellSAT creators explain that “comprehensiveness score captures the extent to which recommended content areas are covered in the policy. The strength score describes how strongly the content is stated. Both scores range from 0 to 100, with lower scores indicating less content and weaker language, and higher scores indicating more content and use of specific and directive language.”²⁵ The scoring tool website clarifies that strong/directive policy language includes concrete plans, timelines and strategies for implementation using enforceable words such as “will, must, require and all,” whereas weaker/less directive language uses vague statements with aspirational goals or recommendations and words such as “may, can, might, try and some.”²⁶

Procedure

A Registered Dietitian Nutritionist with direct experience developing LSWPs scored all policies. A third-year law student and an MPH-educated public health researcher participated in a training session and scored 32 (16%) of the policies to determine interrater reliability. Most intraclass correlations (ICC) scores were 0.8 or higher indicating high agreement between coders. Mean ICC was 0.89 for the total comprehensiveness score and 0.82 for the total strength score. These ICC results were consistent with and slightly higher than those reported by the WellSat creators ²⁵ indicating the interrater reliability was more than adequate.

The adoption date was collected from all policies and classified as “compliant” if adopted after enactment of the USDA final rule on July 26, 2016.¹⁰ Policies adopted prior to July 26, 2016, were considered “non-compliant.”

Adoption of a national or state model LSWP with little or no modification was also determined for each policy using a method employed by Smith et al, ²⁷ identifying if all of the following criteria were present: (1) the name of the model LSWP sponsor was listed on the policy; (2) the policy introduction and subheadings matched the original model, and (3) the number of pages and the organization of the text were similar to the model. Sponsors of the model LSWP were identified as the California School Board Association (classified as CA “state,” ²⁸ the Alliance for a Healthier Generation (classified as “national”),²⁹ and the National Alliance for Nutrition and Activity (classified as “national”).³⁰ We identified 2 different national model LSWPs used in the sample, but the number of LSWPs tied to each was low and combined in the final analysis.

Covariates associated with LSWP quality were identified in the literature including: district size (small < 1000 students, medium 1000 – 5000 students, or large > 5000 students), geographic region (rural/urban), presence (yes/no) of a district high school (since competitive food offerings at high schools require a higher degree of regulation), percent FRPM-eligible students, model policy adoption type (national, state, or none), and policy adoption date (final rule compliant or non-compliant). The sample racial/ethnic data for percent Latino/Hispanic students was highly correlated with percent FRPM eligibility and percent White students (correlation coefficients between 0.54 – 0.79) limiting inclusion of all three variables in the models. Therefore, the district race/ethnicity data were used to create a dichotomized variable (non-White majority or not) to include in the model. Regression coefficients for medium and large district sizes were not statistically different, so these categories were combined into the “large” category.

Data Analysis

The mean comprehensiveness and strength scores with 95% CI were calculated overall and for each covariate category using unadjusted general linear models (GLM) and Tukey’s test to evaluate significant differences in means scores. Two adjusted GLM models were used to examine associations between district and policy-related factors and (1) comprehensiveness and (2) strength scores. For variables with >2 categories, analyses were rerun by changing the referent category to ensure each grouping was compared (not shown in data tables). We tested for interaction between model LSWP type and adoption date using an interaction term (model type x adoption date) because model policies were updated following the 2016 final rule, and those adopted after this date would likely be higher quality than those adopted before. Ten districts missing a policy adoption date were excluded from the final models. Statistical analyses were done using SAS 9.4 (SAS Institute Inc., Cary, NC, USA).

RESULTS

The sample of 200 low-income school districts represent a variety of sizes, locations, and types (Table 1). Mean district enrollment was 10,474 students with a similar proportion of small, medium, and large districts. Almost 70% of districts were in urban areas. About 7% of the districts comprised only high schools (grades 9–12), with the rest as K-12, elementary, or middle school districts.

Table 1.

Characteristics of Low-Income School Districts

District Characteristics (n = 200)	Mean (95% CI)
% FRPM Eligible students a	64.7 (62.0 – 67.3)
District Size a	10,474 (4,121–16,829)
District Size Categories	N (%)
Large (Enrollment ≥ 5000)	72 (36.0)
Medium (Enrollment = 1000 – 4999)	66 (33.0)
Small (Enrollment <1000)	62 (31.0)
Non-White Majority a
Yes	149 (74.5)
No	51 (25.5)
Urbanicity b
Rural	61 (30.5)
Urban	139 (69.5)
District Grade Classification a
Elementary only	76 (38.0)
High School only	14 (7.0)
K - 12	110 (55.0)
Model Policy Type
None	37 (18.5)
State	137 (68.5)
National	26 (13.0)
Policy Adoption Date
Compliant (after Final Rule enacted)	95 (47.5)
Non-Compliant (prior to Final Rule)	95 (47.5)
Missing	10 (5.0)

^aSource: California Department of Education.²²

^bSource: National Center for Education Statistics.²³ Urbanicity is based on standard urban/rural definitions developed by the U.S. Census Bureau. Urban Locales include designations: City, Large; City, Midsize; City, Small; Suburb, Large; Suburb, Midsize; Suburb, Small; Town, Fringe; Town, Distant; Town, Remote. Rural Locales include designations: Rural, Fringe; Rural, Distant; Rural, Remote.²⁴

Because our sample included districts with low-income schools, the poverty indicator was high with nearly 65% of students FRPM-eligible. Districts represented a racially and ethnically diverse student population (75% had a Non-White majority). Half of the policies were adopted before the 2016 final rule mandate. Most districts (68.5%) adopted the state model LSWP. Another 13% adopted a national model, and 18.5% created an original policy.

The mean WellSAT scores were 65.0 out of 100 for comprehensiveness and 37.3 out of 100 for strength (Table 2). There were no significant differences in mean comprehensiveness or strength scores by district size or geographic region. Mean comprehensiveness score for districts with at least one high school was similar to those without high schools but mean strength score was 4 points higher for districts with high schools (p = 0.048). Mean comprehensiveness scores for districts adopting the state model policy were 4.2 points higher (p = 0.043) and those adopting a national model were 18.2 points higher (p = 0.001) than districts that developed an original LSWP. Strength scores were also significantly higher with model policy adoption. Districts adopting the state model had a 4.7 point higher mean score (p=0.046) and districts adopting a national model had a 21.8 point higher mean score (p<0.001), than those with an original LSWP. There was no difference in mean comprehensiveness score by policy adoption date, but the mean strength score for date-compliant policies was 9.4 points higher than for districts with non-compliant policies (p<0.001).

Table 2.

Mean Strength and Comprehensiveness Scores by District and Policy Characteristics

	Unadjusted Score Mean (95% CI)
District Characteristics	Comprehensiveness score	Strength score
Overall Mean (n=200)	64.95 (63.23 – 66.66)	37.29 (35.34 – 39.24)
District size
Large	65.07 (62.20 – 67.94)	37.61 (34.36 – 40.87)
Medium	65.83 (62.83 – 68.83)	38.49 (35.09 – 41.89)
Small (ref)	63.86 (60.77 – 66.96)	35.63 (32.12 – 39.13)
Non-White Majority
Yes	65.06 (63.06 – 67.05)	36.63 (34.37 – 38.89)
No	64.62 (61.21 – 68.03)	39.21 (35.36 – 43.07)
Geographic Region
Rural	65.23 (62.18 – 68.35)	37.28 (33.74 – 40.81)
Urban	64.82 (62.76 – 66.89)	37.29 (34.95 – 39.64)
District has High School(s)
No	63.60 (60.81 – 66.38)	34.80 (31.18 – 38.41) *
Yes	65.78 (63.60 – 67.95)	38.81 (36.58 – 41.05)
Model Policy Type
National	77.80 (73.47 – 82.13) *	53.08 (48.24 – 57.92) *
State	63.92 (62.04 – 65.81) *+	35.92 (33.81 – 38.03) *+
None (ref)	59.70 (56.07 – 63.34)	31.26 (27.21 – 35.32)
Policy Adoption Date
Compliant	66.11 (63.62 – 68.61)	41.83 (39.15 – 44.51) *
Non-Compliant	63.75 (61.26 – 66.25)	32.44 (29.76 – 35.13)

^*p-value <0.05 using Tukey’s test to evaluate differences in means.

⁺Significantly different from the national template.

Results for the adjusted linear regression for comprehensiveness scores are presented in Table 3. The model LSWP type x policy date interaction term was not significant (p=0.11) and was excluded from the final model. The R² was 0.21 meaning this model explained 21% of the variation in comprehensiveness scores. Comprehensiveness scores were 5.1 points higher in large districts compared with small (p = 0.024). Districts adopting a national model LSWP scored 17.7 points higher than districts with original LSWPs (p < 0.0001) and 15.3 (95% CI: 10.2, 20.4) points higher than those using the state model (p <0.0001) (results not shown in table). Mean comprehensiveness for districts that developed their own LSWP were not statistically different than those adopting the state model LSWP (results not shown in table).

Table 3.

Linear Regression Results for Comprehensiveness Score

District Characteristics (n=190)	Adjusted estimate*	95% CI	p-value
District Size
Large (Enrollment ≥ 1,000)	5.12	0.69 – 9.57	0.024
Small (Enrollment < 1,000)	ref	---	---
Non-White Majority
No	−2.31	−6.21 – 1.59	0.244
Yes	ref
% Students Eligible for FRPM	−0.05	− 0.14 – 0.04	0.310
Urbanicity
Rural	1.63	−8.23	0.435
Urban	ref	---	---
District has High School(s)
No	−0.89	− 4.22 – 2.44	0.601
Yes	ref	---	---
Model Policy Type
National	17.66	11.58 – 23.75	<0.0001
State	2.38	− 1.74 – 6.50	0.255
None	ref	---	---
Policy Adoption Date
Compliant	1.55	−1.55 – 4.67	0.325
Non-Compliant	ref		---

^*Adjusted for district size, non-White majority, % students eligible for FRPM, geographic region, district has high school(s), template type, and adoption date.

ref = reference group

Results for the adjusted linear regression for strength scores are presented in Table 4. The model LSPW type x date interaction term was statistically significant (p=0.024), and results are presented for districts by compliant and non-compliant adoption date. The R² of 0.36 for the date compliant model means it explains 36% of the variation in strength scores. The R² of 0.18 for the non-compliant model means it explains 18% of the variation in strength scores.

Table 4.

Linear Regression Results for Strength Score, Stratified by Policy Adoption Date

District Characteristics	Adoption Date Compliant (n=95)			Adoption Date Non-Compliant (n= 95)
	Adjusted* estimate	95% CI	p-value	Adjusted* estimate	95% CI	p-value
District Size
Large (Enrollment ≥ 1,000)	7.22	0.69 – 13.75	0.031	6.13	−14.87	0.105
Small (Enrollment < 1,000)	ref	---	---	ref	---	---
Non-White Majority
No	−1.32	−7.23 – 4.58	0.657	0.43	−5.81 – 6.68	0.890
Yes	ref
% Students Eligible for FRPM	−0.03	−0.17 – 0.11	0.665	−0.13	−0.38	0.073
Urbanicity
Rural	0.68	−12.44	0.829	4.21	−2.29 – 10.71	0.202
Urban	ref	---	---	ref	---	---
District has High School(s)
No	−0.42	−5.37 – 4.54	0.868	−4.22	−9.92 – 1.48	0.145
Yes	ref	---	---	ref	---	---
Model Policy Type
National	26.99	17.64 – 36.35	<0.0001	11.73	1.59 – 21.88	0.024
State	4.17	− 3.12 – 11.46	0.259	0.43	−5.10 – 5.96	0.877
None	ref	---	---	ref	---	---

^*Adjusted for district size, non-White majority, % students eligible for FRPM, geographic region, district has high school(s), template type, and adoption date.

ref = reference group

For the districts with date-compliant policies, strength score was 7.2 points higher in large compared to small districts (p=0.031). For non-compliant districts, the strength score was 6.1 points higher for large districts, but the difference was not statistically significant (p=0.105). The districts with date-compliant policies that adopted a national model LSWP scored 27.0 points higher than those that developed an original policy (p < 0.001). This estimate is more than double the estimate for districts adopting the national model with non-compliant adoption dates. For date compliant policies, the mean strength for adoption of a national model LSWP was also significantly higher than for adoption of the state model (Estimate = 22.8, [15.9, 29.8], p < 0.001) (results not shown in table). For date non-compliant policies, the mean strength score for adopting the national model policy was also significantly higher than for adoption of the state model (Estimate = 11.3, [1.8, 20.8], p = 0.021). There was no difference in strength score for districts adopting the state model LSWP and those that developed an original policy (results not shown in table).

DISCUSSION

All districts in this sample of low-income-serving CA public districts had adopted a LSWP by 2018. Mean policy quality scores were similar to national scores; the mean comprehensiveness score was 65 (vs. 53), and the mean strength score was 37 (vs. 33).¹⁸ Adoption of model LSWPs, near verbatim, was common (> 80% of districts). Districts adopting a national model LSWP had significantly higher comprehensiveness and strength scores than those that adopted the state model policy and those that developed an original policy. Quality scores for districts that adopted the state model were not significantly different from districts that developed an original LSWP. Districts size (≥ 1,000 students) was also associated with higher policy comprehensiveness, and adoption after the USDA final rule was associated with higher policy strength.

The high adoption of model LSWPs, in this sample, is comparable to results from LSWP evaluations conducted in other states,^{27 31 32 33} and supports findings that model policy adoption may influence LSWP quality scores, for better or worse. In our study, the state model LSWP was adopted most often (by more than 2/3 of districts), and a national model was adopted by only 13% of districts. The small number of districts adopting a higher quality national policy in our sample limited further analysis of these districts.

Districts may have a higher degree of trust and comfort with guidance from known local school-based organizations, which could explain the preference for the state model LSWP in our sample. Furthermore, the national model LSWPs are more comprehensive in scope and incorporate a greater degree of directive language for both required elements and those recommended as best practices, resulting in higher policy comprehensiveness and strength. The less extensive state model LSWP incorporates strong language for required policy elements, but less directive or no language for recommended LSWP elements resulting in lower policy quality scores. Some districts may not feel the “recommended” elements are necessary or may find difficulty committing to long-term implementation of these elements. However, the recommended LSWP elements often provide evidence-based strategies that support implementation of required policy elements, without which the overall impact on student health may be reduced. A recent systematic review of strategies to increase consumption of nutritious school meal offerings showed several recommended LSWP school meal elements such as adequate seat time, limiting access to competitive foods and promotion of healthy foods at lunch were clearly associated with higher rates of school meal consumption.³⁴

LSWPs for this study were collected in spring 2018 to reflect policies developed after the 2016 final rule mandating policy revision and updates by June 30, 2017. However, half of the sample policies were adopted prior to the enactment of this mandate without evidence of update compliance. Higher strength scores among policies adopted after the final rule may suggest these districts reviewed updated LSWP expectations and made appropriate revisions, or at least used a model LSWP compliant with the mandate. While low LSWP update compliance was also common in evaluations from other states,³⁵ it’s unclear whether this stems from lack of knowledge about required updates or lack of interest in LSWP development. Combined, these findings suggest promoting a stronger model LSWP by a trusted source and regular policy revision may improve LSWP quality and, ultimately, the school food and physical activity environment. Evidence also suggests that additional state-level LSWP mandates and oversight are associated with higher LSWP quality and implementation.³⁶

We did not find associations between LSWP quality and percent FRPM-eligible students, non-White majority districts, urbanicity or having high school(s). Some studies found higher LSWP scores associated with serving large numbers of low-income and/or racial/ethnic minority students (likely because higher FRPM participation rates often require stricter accountability to school food regulations). The high proportion of FRPM-eligible and non-White majority districts in our sample may explain why no association was found for these variables.

Implications for School Health Policy, Practice, and Equity

Several studies propose explanations for low LSWP engagement and suggestions to manage high model LSWP adoption. Lucarelli et al. explain that districts may “intentionally keep written policies vague so that each building can tailor the policy to their specific needs or for fear of auditing of wellness practices,” also noting that policies are commonly “accompanied by a procedure manual (which may be deemed less restrictive for districts in which a lengthy and expensive process is required to change policy-related documents) outlining more specific requirements to implement the policy.”³² Szeszulski et al. suggest adding intentional spaces in model LSWPs where districts add details aligned to specific local priorities may increase engagement in policy development and improve quality.³³ Perspectives from focus groups and key informant interviews with school administrators consistently report lack of time and resources limit commitment to LSWP development and implementation due to significant and frequently competing high-priority demands in the school environment and a higher accountability for academic outcomes.³⁷ School authorities commonly recommend allocation of federal and state funding for dedicated district level personnel to coordinate health programs and policies to support further LSWP improvements and implementation.^{37 38} Personal experience in this field substantiates these challenges along with findings that some districts may choose weaker, less directive policy language to manage accountability expectations given the fluctuating circumstances and tight resource environments common to the school setting.

The recent COVID-19 pandemic school closures further tested district capacity to manage student wellness and food and activity environments by requiring districts to both educate and feed students regardless of whether they were on campus or at home. Frequent, sometimes daily updates to local, state, and national education and school meal policies and practices were needed to manage rapidly fluctuating circumstances.³⁹ Future LSWP development should consider contingency plans to manage LSWP provisions and implementation during local and national crises and/or school closures which are likely to persist given the growing frequency of both climate and public health emergencies.

Recommendations to improve LSWP engagement from the school health literature suggest that district-wide adoption of a comprehensive and integrated structure to organize school wellness such as the CDC’s Whole School, Whole Community, Whole Child (WSCC) school health model, could provide a useful framework for LSWP development by promoting cooperation between academic, health and nutrition units within districts.^{40 41} The WSCC emphasizes relationships between school departments and community partnerships to broaden school district objectives beyond simply academic goals to “incorporate improvements that address barriers to learning such as health and well-being and provide a structure for collaboration among health and education sectors at the local and state level.” ⁴¹ LSWP elements align with several WSCC model domains. The San Diego Unified School District collaborated with community partners using this approach to guide a district wide LSWP review tied to the WSCC model and significantly improved the quality and implementation of their LSWP.⁴²

Murray et al.³⁸ also advocate for linking LSWP components to a district’s local control funding formula (LCFF) and school climate framework which can broaden support and leverage resources to improve both student health and academic outcomes. Further research into this approach is needed, including focus groups and key informant interviews with successful districts to inform further use in LSWP development.

Limitations

The study’s focus on low-income CA public districts limits generalization of findings to all CA districts and those outside the state. High correlation between district racial/ethnic makeup and FRPM eligibility in the study sample limits examination of influences from individual racial/ethnic variables on LSWP scores. However, the percent FRPM eligibility and districts with a non-White majority variables included in the analysis showed no association with LSWP quality indicating individual racial/ethnic variables are not likely associated with LSWP quality among these districts. Another study limitation is the single policy sampling period from districts’ public websites. We were unable to determine if district school wellness websites were up to date, and newer policies may have been adopted since this evaluation sampling period. Ten policies did not include an adoption date and could not be included in the analysis.

CONCLUSIONS

LSWP mandates have resulted in significant improvements in school food and physical activity environments that promote student health and academic success. While strong and comprehensive LSWPs show commitment and support implementation and accountability on the ground, study findings that most district policies simply adopted a model LSWP may indicate that districts do not have the interest or bandwidth to develop LSWP quality beyond meeting basic requirements.

The recent COVID pandemic demonstrated that schools are an important partner in ensuring access to nutritious food and physical activity that support good health. However, poor nutrition and inactivity are not the only child health threats our nation’s schools must manage daily. As districts move forward in the post-COVID era, more efficient and flexible LSWP processes and expectations may be necessary to support student health. Further research is needed including focus groups and key informant interviews with school administrators and school food authorities to improve understanding about how to use LSWP development more effectively to support student wellness. In-depth evaluations of districts with high quality wellness policies and nutrition programs may be useful to fill gaps in knowledge and guide future efforts to support integration of LSWPs within other priority school health and education frameworks such as the WSCC and LCFFs to leverage limited district resources more efficiently to promote both student health and academic success.

Human Subject Approval Statement.

Preparation of this paper did not involve data collection involving human subjects, and therefore, no institutional review board examination or approval was required.