Social Determinants of Health Correlate With Lower CRC Screening Rates


Isabelle Franklin discusses the effects of social determinants of health on rates of colorectal cancer screening, expands on how these social determinants of health could affect a patient’s likelihood of adhering to screening guidelines, and highlights ways to potentially address these barriers to screening.

Isabelle Franklin

Isabelle Franklin

Financial strain, social isolation, and food insecurity had statistically significant associations with lower rates of colorectal cancer (CRC) screening, according to a cross-sectional study of 3443 patients in the Kaiser Permanente health-care system across regions including California, Colorado, Washington, Georgia, and Hawaii.

Data from the cross-sectional study presented at the 2023 ASCO Annual Meeting showed that among patients without severe financial strain, 83% (n = 2457) were up to date on their CRC screening in 2020. Among those who reported severe financial strain, 74% (n = 325) received CRC screening in 2020.

In patients who were not socially isolated, 82% (n = 2514) were screened; however, only 76% of patients (n = 286) who reported social isolation underwent screening. Additionally, 82% of patients (n = 2679) without food insecurity received CRC screening; 71% of patients (n = 107) who reported food insecurity underwent screening.

Additional findings demonstrated that demographic factors such as age, insurance status, and a primary language other than English were found to be significantly correlated with CRC screening rates, as patients were less likely to be screened if they were younger, healthier, and individually insured.

“The data show that even in an integrated system like Kaiser Permanente, in which all of our members are insured and have health-care coverage, social determinants remain drivers affecting health,” lead study author Isabelle Franklin said in an interview with OncLive®. “Ultimately, health-care systems need to be investing in programs to help mitigate these social needs if they want to increase their cancer screening rates and improve the health of their population.”

In the interview, Franklin, a medical student at Kaiser Permanente Bernard J. Tyson School of Medicine, in Pasadena, California, discussed the effects of social determinants of health on rates of CRC screening, expanded on how these social determinants of health could affect a patient’s likelihood of adhering to screening guidelines, and highlighted ways to potentially address these barriers to screening.

OncLive: What the rationale for investigating the relationship between social determinants of health and CRC screening rates?

Franklin: This study looked at the correlation between CRC screening and patients’ self-reported social needs in the categories of financial strain, housing instability, transportation issues, social isolation, and food insecurity. All [patients were] within the Kaiser Permanente patient population.

For a little background, CRC is one of the leading causes of cancer death in the United States. However, CRC screening has consistently lagged behind similar cancer screening rates, such as breast and cervical cancer. Many studies have previously been conducted to attempt to identify the driving factors behind these lower screening rates, and several large, systematic reviews of patient-reported barriers to CRC screening have previously been conducted. Social determinants of health have emerged as potentially some of those principle driving factors behind low screening rates. Due to their modifiable nature, [social determinants of health] are an ideal target for quality improvement–focused research, such as this.

[Although] patient-reported barriers have previously been described, fewer studies have analyzed how those patient-reported barriers and social needs affect screening rates. Our study is unique in that due to Kaiser Permanente’s robust electronic health records, we have access to so much data. We were able to link those patient-reported survey data and social needs to their electronic medical record screening status to give us a good idea of what patient-reported needs and social needs are linked to lower screening rates.

The overall hope for this research is to inform future quality improvement efforts within the Kaiser Permanente region to hopefully increase our cancer screening rates.

What methods were utilized in this study?

In terms of the patient population, our population included Kaiser Permanente members, and all those are members who are fully insured. The population’s age range was 50 to 75 [years at the time] the social needs survey was sent out in 2020. That was the United States Preventive Services Task Force screening age range at that time. Now, the age range has been lowered to 45 to 75 [years].

Overall, this was a cross-sectional study that merged data from Kaiser Permanente's 2020 national social needs survey with the survey respondents’ electronic health record CRC screening data. The primary outcome for the analysis was being up to date with CRC screening [information] in the year 2020. The primary measures looked at were those 5 social needs within the survey: financial strain, food insecurity, housing instability, transportation issues, and social isolation. Those needs were correlated with each patient’s screening status to attempt to define patient-level characteristics that could be predictors of low screening rates, and, ultimately, odds ratios were calculated.

Could you expand on the key findings from this study presented at the 2023 ASCO Annual Meeting?

When overlaying the social needs and the screening completion data, we ultimately found significant correlation between 3 of those 5 social needs and lower rates of screening. Those 3 domains were financial strain, social isolation, and food insecurity. However, even those other 2 domains—transportation [issues] and housing [instability]—were borderline significant.

The data we presented at the 2023 ASCO Annual Meeting was an aggregate of all patients from across all of the regions within Kaiser Permanente. However, it's important to know that, internally, Kaiser Permanente is broken up into various regions. An important note is that each of the 7 regions that we looked at, when looked at in isolation, did have slightly different results on which of the 5 social needs were significant. Transportation and housing, though not significant overall, were significant when looked at in some of those regions specifically. That's important just for the overall takeaway. All of these social needs could potentially be playing a role in affecting screening rates, and each of them could be possible useful targets for future quality-improvement efforts.

These results are particularly striking, seeing as the study was conducted within the Kaiser Permanente system, in which all patients are insured. Many people might just assume that having health insurance negates all problems [with cancer screening]. Unfortunately, our data show that these disparities can persist, and there is more work to be done in addressing them.

How can health-care providers work to address or mitigate some of these disparities and needs in their own practices going forward?

The first step is having mechanisms in place to make sure you're asking patients about these social determinants, whether that’s with an intake screening survey when patients come into clinic or making sure providers are actually asking their patients about these things when they're in their visits with them. It's crucial to know where your patients are coming from, and you can't address these disparities unless you know about them in the first place.

From there, once you've identified a need, the more challenging task can be addressing it. This is where providers being familiar with the organizations, the resources, and the community around them can be very helpful so that they can know where to refer their patients to get help with the needs, be it food, housing, etc.

Kaiser Permanente has a new program that is piloting in some of its regions right now currently called Thrive Local, which is a database of hundreds of community resources and organizations that can be filtered by type of social need and by a patient's zip code. It is integrated within the electronic health records so that providers can refer their patients to these organizations directly through their patient portal and even get notifications when their patients have been connected to those resources. That is a cool model for addressing patient needs, and the ability for providers to be able to get real-time data back of when their patients are being connected to these resources is an interesting step in ensuring that they're getting the help that they need.

Are there any next steps for this research that you would like to highlight?

On our side of things, I've been collaborating with quality-improvement leaders within each of the Kaiser Permanente regions since the initiation of this project. My team is now in the process of setting up meetings with each of those regional leaders to share their region’s specific data with them. The hope is that they'll be able to take those results back to their teams and hopefully they'll inform the quality-improvement initiatives that they'll be implementing to increase screening rates within Kaiser Permanente.

On a larger scale, the biggest implication of this type of research is that it hopefully reinforces the importance of screening for social needs and encourages, at a system level, investing in programs like Thrive Local that leverage community organizations to get patients connected to the resources that they need.


Franklin I, Gambatese R, Duggan M, et al. The effects of social determinants of health on colorectal cancer screening. J Clin Oncol. 2023;41(suppl 16):6568. doi:10.1200/JCO.2023.41.16_suppl.6568.

Related Videos
Chun Chao, PhD, MS
Efficacy and Safety of Fruquintinib in Patients With Metastatic Colorectal Cancer According to Prior Treatment Sequence in the Refractory Setting: Results From FRESCO and FRESCO-2
Richard Kim, MD
Salman R. Punekar, MD
Chun Chao, PhD, MS
Christos Kyriakopoulos, MD
Marwan G. Fakih, MD
John L. Marshall, MD
Rene Adam, MD, PhD
Amin Nassar, MD