Socioecological Determinants of Health and the Quality of Colonoscopy in Rural Alabama

Abstract

BACKGROUND:

Rural patients experience a higher incidence of and mortality from colorectal cancer. Ensuring high-quality screening is essential to address these disparities.

OBJECTIVE:

To investigate whether socioecological determinants of health are associated with colonoscopy quality in rural Alabama.

DESIGN:

Retrospective review.

SETTING:

Data across 3 rural hospitals in Alabama from August 2021 to July 2023.

PATIENTS:

We included adults (aged 18 years or older) who underwent screening or diagnostic colonoscopy and completed a validated survey that measures socioecological determinants of health.

MAIN OUTCOME MEASURES:

Primary outcomes included bowel preparation quality, cecal intubation, and adenoma detection rate. We linked the survey responses to these quality metrics to identify factors associated with outcomes. Analyses included the χ², Fisher exact, and Kruskal-Wallis rank-sum tests, with a p value of < 0.05 considered statistically significant.

RESULTS:

The 84 patients surveyed were 66.7% men, 50.0% Black, and had a median age of 64 years. Optimal bowel preparation was present in 88.0%, successful cecal intubation was observed in 89.3%, and the overall adenoma detection rate was 45.8%. Patients with suboptimal bowel preparation described lower rates of internet access (60.0% vs 87.4%, p < 0.05), more difficulty in understanding written information (30.0% vs 1.4%, p < 0.05), and lacked a sense of responsibility for their health (30.0% vs 51.4%, p < 0.05) compared to those having optimal bowel preparation. Those with unsuccessful cecal intubations had lower physician trust (55.6% vs 73.3%, p < 0.05), whereas patients with successful cecal intubations were more confident in preventing health-related problems (53.3% vs 33.3%, p < 0.05) and had a more supportive social environment (72.0% vs 66.7%, p < 0.05).

LIMITATIONS:

Retrospective design and small sample size limiting multivariable analyses.

CONCLUSION:

In rural Alabama, lower health literacy, internet access, and physician trust were associated with low-quality colonoscopy, whereas a higher patient sense of responsibility and a supportive social environment were associated with higher-quality metrics. These findings identify potential targets for improving colonoscopy quality in rural settings. See Video Abstract.

Colorectal cancer (CRC) is the third most common cancer affecting both men and women in the United States and is the second leading cause of cancer-related deaths.¹ Although the incidence of CRC has steadily decreased over the past several decades,² this decrease has not been equally experienced across different regions, including in rural settings. Extant literature has shown that rural patients had disproportionately higher incidence,² more advanced stages at diagnosis,³ and higher mortality of CRC compared to urban patients.⁴ Therefore, ensuring high-quality cancer screening is essential to address these disparities and improve outcomes,⁵ particularly in states such as Alabama, where 43.6% of the population resides in rural areas⁶ and which has the largest rural-to-urban CRC incidence disparity in the United States.⁷

Colonoscopies play a critical role in reducing CRC disparities. Their role in CRC screening was supported by findings from the National Polyp Study, which demonstrated a 76% to 90% reduction in CRC incidence and a 53% reduction in mortality among patients who had a colonoscopy and polypectomy.⁸ As such, colonoscopy is the most commonly performed screening test for CRC in the United States.⁹ The overall effectiveness, however, is dependent on the quality of the procedure, which is quantified by metrics such as bowel preparation quality, successful cecal intubation with photographic documentation, and adenoma detection rate (ADR).¹⁰ Unfortunately, suboptimal bowel preparation remains a common occurrence, affecting more than 20% of all colonoscopies.¹¹ This results in lower ADRs,¹¹ an increased need for repeat colonoscopy within a time frame discordant with the guidelines,¹² and ultimately increased health care costs¹³ and risk of procedure-related complications.¹⁴

Despite this common occurrence, a knowledge gap exists regarding predictors of low-quality colonoscopy. Only a few individual-level factors have been linked to colonoscopy quality, including advanced age,¹⁵ male sex,¹⁶ and comorbidities.¹⁷ The influence of modifiable socioecological determinants of health (SEDOHs) on colonoscopy quality is particularly poorly studied as prior studies have focused on a limited number of variables such as insurance and marital status.¹⁶ Given this, we aimed to perform an exploratory analysis investigating whether SEDOHs are associated with the quality of colonoscopy in rural Alabama. Identification of modifiable SEDOHs associated with colonoscopy quality would inform strategies to improve the quality of colonoscopies in these low-resourced rural settings.

MATERIALS AND METHODS

Study Design, Inclusion Criteria, and Exclusion Criteria

This study was a retrospective review of a prospectively maintained database involving surgical patients seen between August 2021 and July 2023 from 3 rural hospitals in Alabama: Whitfield Regional Medical Center in Demopolis, Russell Medical Center in Alexander City, and Regional Medical Center of Central Alabama in Greenville. We included adults (aged 18 years or older) who were English-speaking, had undergone a screening or diagnostic colonoscopy at one of the 3 sites during the study period, and were able to give consent. Included participants completed the SEDOH-88¹⁸ survey. We excluded patients who were non-English speakers, were younger than 18 years, were unable to give consent, and underwent a procedure other than a colonoscopy. The study protocol and supplemental materials were approved by the University of Alabama at Birmingham (UAB) Institutional Review Board as part of the Deep South Resource Center for Minority Aging Research project (IRB-300006942). The Strengthening the Reporting of Observational Studies in Epidemiology guidelines were used for study reporting.¹⁹

Study Sites (Setting)

Two of the 3 rural health care facilities (Demopolis in Marengo County and Greenville in Butler County) are located within the Black Belt of Alabama, and Alexander City in Tallapoosa County is within close proximity. Formerly named for the color of its fertile soil and later for the large proportion of Black residents, the Black Belt has distinct geographic, social, and cultural boundaries that underpin the high rates of poverty and health care disparities in the region.²⁰

According to the 2020 to 2022 US Census Bureau population estimates,²¹ Marengo, Butler, and Tallapoosa County comprised 18,745, 18,650, and 40,977 residents, respectively. More granularly, in 2019, the Black Belt comprised 56% Black residents, with nearly 1 in 4 individuals (23.7%) living below the poverty line.²⁰ Furthermore, according to the 2020 U.S. Centers for Disease Control and Prevention statistics,²² the nationwide social vulnerability index for Marengo, Butler, and Tallapoosa County was 0.81, 0.68, and 0.89, respectively. With social vulnerability index scores approaching 1 indicative of higher social vulnerability, these findings highlight the degree of destitution within these rural communities. These 3 rural health care facilities were selected as they form part of the UAB Surgery Community Network.²³ All colonoscopy procedures included in this analysis were performed by UAB general surgeons who are positioned full time at each rural site.

Survey Design and Collection

The SEDOH-88¹⁸ is an 88-item survey that captures individual, structural, and environmental health-related factors across multiple socioecological levels,²⁴ including the individual (health literacy), interpersonal (physician trust), organizational (health care cost and insurance), and community domains (health care access and community support; Fig. 1). The survey was previously developed using a 2-phase modified-Delphi method by a group of experts in surgery, sociology, and social determinants of health through a collaborative effort between the UAB Department of Surgery, the UAB Minority Health and Health Disparities Research Center, and the Social Determinants of Health Core, respectively.

FIGURE 1.

SEDOH-88 survey components mapped to the socioecological model. Adapted from Smith et al.¹⁸ SEDOH = socioecological determinants of health.

As part of an ongoing prospective study, patients undergoing general surgical procedures, including screening and diagnostic colonoscopy procedures at each of the 3 sites, were approached either in-person during preoperative encounters or by phone for completion of informed consent and survey completion. For in-person encounters, the survey was verbally delivered by study personnel, and answers were recorded on an iPad device directly into the Deep South Surgical Equity Research Network Research Electronic Data Capture database hosted and coordinated by UAB Research Electronic Data Capture servers. For phone surveys, patients were considered for inclusion if they had a colonoscopy within 6 months from the time of survey administration. Each survey was approximately 12 to 15 minutes in length. For this study, only patients undergoing colonoscopy procedures were included.

Outcomes and Independent Variables

The primary outcomes were standardized colonoscopy quality metrics potentially impacted by patient-level SEDOH measures, including bowel preparation quality, successful cecal intubation, and ADR. Withdrawal time was not analyzed because this information was not available in the majority of patients and was less likely to be dependent on patient-level SEDOHs. Based on the Aronchick scale, the quality of bowel preparation was dichotomized into “optimal” (a composite of “excellent,” “good,” and “fair” preparation) or “suboptimal” (a composite of “poor” and “inadequate” preparation).¹¹ Cecal intubation was classified as “successful” versus “unsuccessful” based on endoscopic visualization of the cecum, appendix orifice, or ileocecal valve. ADRs were dichotomized as “adenoma(s) detected” or “no adenoma detected.” Colonoscopy quality metrics were abstracted from the medical records of patients at each site, uploaded to the Deep South Surgical Equity Research Network database, and linked to the SEDOH-88 responses.

Data Analysis

Descriptive statistics were conducted using frequencies and percentages for categorical variables, whereas continuous variables were summarized with means and SD or medians with interquartile values. Bivariate analyses compared the quality metrics of colonoscopy by the SEDOH-88 survey responses using the χ² or Fisher exact probability test as appropriate for categorical variables and the Kruskal-Wallis rank-sum test for continuous variables. A p value of <0.05 was considered statistically significant. Data analysis was performed using SAS software version 9.4 (SAS Institute, Cary, NC).

RESULTS

Demographics

Of the 178 surgical patients contacted, 155 underwent a colonoscopy. Of these, 84 (54.2%) agreed to participate in the study. Twenty-five patients completed the SEDOH-88 survey in-person, whereas an additional 59 completed the survey via phone. Included participants’ SEDOH-88 responses and colonoscopy quality metrics were analyzed. A majority (n = 67; 79.8%) underwent a screening colonoscopy, whereas 17 (20.2%) had a diagnostic colonoscopy. A similar number of participants were recruited from each site: 30 (35.7%) from Demopolis, 27 (32.1%) from Alexander City, and 27 (32.1%) from Greenville. The overall median age of participants was 64 years (interquartile range, 54.5–70.0), 66.7% were men, and 50.0% (n = 42) were of Black race (Table 1).

TABLE 1.

Demographics and colonoscopy quality metrics (N = 84)

Variable	Overall (N = 84)
Age, y, median (IQR)	64 (54.5–70)
Sex, n (%)
Male	56 (66.7)
Female	28 (33.3)
Race, n (%)
White	41 (48.8)
Black	42 (50.0)
Asian	1 (1.2)
Bowel preparation quality, n (%)
Excellent	27 (32.1)
Good	8 (9.5)
Fair	39 (46.4)
Poor	4 (4.8)
Inadequate	6 (7.1)
Successful cecal intubation, n (%)
Yes	75 (89.3)
No	9 (10.7)
No. of polyps excised, n (%)
0	45 (54.2)
1–2	25 (30.1)
3–4	6 (7.2)
≥5	7 (8.4)
Withdrawal time, n (%)
<6 min	1 (1.2)
>6 min	42 (50.0)
Not recorded	41 (48.8)

Colonoscopy Quality Metrics

Optimal bowel preparation (“excellent,” “good,” or “fair” on the Aronchick scale) was present in 88.0% of cases (Table 1). We observed a cecal intubation rate of 89.3%. The overall ADR was 45.8%. Of these, the majority (30.1%) had 1 to 2 polyps detected and excised.

SEDOHs by Colonoscopy Quality Metrics

Optimal vs suboptimal bowel preparation.

Patients with suboptimal bowel preparation described lower rates of internet access (60.0% vs 87.4%; p = 0.02), had more difficulty in understanding written medical information (30.0% vs 1.35%; p = 0.002), had more often required help reading hospital materials (30.0% vs 2.7%; p = 0.007), and had more challenges in taking responsibility for their health (30.0% vs 51.4%; p = 0.010) compared to those having optimal bowel preparation (Table 2). Patients with suboptimal bowel preparation were more likely to be Medicaid enrollees compared to those with optimal bowel preparation (30.0% vs 6.8%; p = 0.038).

TABLE 2.

Comparison of SEDOHs by optimal vs suboptimal bowel preparation (N = 84)

Variable	Overall (N = 84)	Optimal bowel preparation (N = 74)	Suboptimal bowel preparation (N = 10)	p
Employment status				0.021
Employed	30 (35.7)	28 (37.8)	2 (20.0)
Unemployed	1 (1.2)	1 (1.4)	0 (0.0)
Retired	30 (35.7)	28 (37.8)	2 (20.0)
Student	1 (1.2)	0 (0.0)	1 (1.0)
Disabled	19 (22.6)	14 (18.9)	5 (50.0)
Other	3 (3.6)	3 (4.1)	0 (0.0)
Do you have difficulty understanding written medical information?				0.002
Always	4 (4.8)	1 (1.4)	3 (30.0)
Often	3 (3.6)	3 (4.1)	0 (0.0)
Sometimes	9 (10.7)	9 (12.2)	0 (0.0)
Occasionally	8 (9.5)	7 (9.5)	1 (10.0)
Never	60 (71.4)	54 (73.0)	6 (60.0)
How often do you have someone help you read the hospital materials?				0.007
Always	9 (10.7)	7 (9.5)	2 (20.0)
Often	5 (6.0)	2 (2.7)	3 (30.0)
Sometimes	10 (11.9)	9 (12.2)	1 (10.0)
Occasionally	10 (11.9)	10 (13.5)	0 (0.0)
Never	50 (59.5)	46 (62.2)	4 (40.0)
I am the person responsible for taking care of my health.				0.012
Strongly disagree	1 (1.2)	0 (0.0)	1 (10.0)
Disagree	2 (2.4)	1 (1.4)	1 (10.0)
Agree	40 (47.6)	35 (47.3)	5 (50.0)
Strongly agree	41 (48.8)	38 (51.4)	3 (30.0)
I am confident that I can help prevent or reduce problems associated with my health.				0.027
Disagree	4 (4.8)	3 (4.1)	1 (1.0)
Agree	43 (51.2)	40 (54.1)	3 (30.0)
Strongly agree	36 (42.9)	31 (41.9)	5 (50.0)
NA	1 (1.2)	0 (0.0)	1 (1.0)
Do you have internet access?				0.022
Yes	71 (84.5)	65 (87.8)	6 (60.0)
No	13 (15.5)	9 (12.2)	4 (40.0)
If you needed it, how often is someone available to turn to for suggestions about how to deal with a personal problem?				0.690
None of the time	1 (1.2)	1 (1.4)	0 (0.0)
A little of the time	4 (4.8)	3 (4.1)	1 (10.0)
Some of the time	9 (10.7)	9 (12.2)	0 (0.0)
Most of the time	12 (14.3)	10 (13.5)	2 (20.0)
All of the time	58 (69.0)	51 (68.9)	7 (70.0)
If you needed it, how often is someone available who understands your problems?				0.440
A little of the time	5 (6.0)	5 (6.8)	0 (0.0)
Some of the time	8 (9.5)	8 (10.8)	0 (0.0)
Most of the time	15 (17.9)	12 (16.2)	3 (30.0)
All of the time	56 (66.7)	49 (66.2)	7 (70.0)
Your doctor only thinks about what is best for you.				0.160
Strongly disagree	1 (1.2)	1 (1.4)	0 (0.0)
Disagree	1 (1.2)	0 (0.0)	1 (10.0)
Neutral	7 (8.3)	6 (8.0)	1 (10.0)
Agree	60 (71.4)	53 (71.6)	7 (70.0)
Strongly agree	14 (16.7)	13 (17.6)	1 (10.0)
Unsure	1 (1.2)	1 (1.4)	0 (0.0)
Are you currently covered by any of these insurance plans?				0.038
Medicare	31 (36.9)	26 (35.1)	5 (50.0)
Medicaid	8 (9.5)	5 (6.8)	3 (30.0)
Blue Cross/Blue Shield of Alabama	28 (33.3)	26 (35.1)	2 (20.0)
Other private insurance	17 (20.2)	17 (23.0)	0 (0.0)
How easy is it for you to make an appointment if you are sick and need health care?				0.710
Very easy	50 (59.5)	43 (58.1)	7 (70.0)
Easy	32 (38.1)	29 (39.2)	3 (30.0)
Neither easy nor difficult	2 (2.4)	2 (2.7)	0 (0.0)
How long has it been since you last saw a health care professional about your health?				0.006
Within the past year	83 (98.8)	74 (100.0)	9 (90.0)
Never or unsure	1 (1.2)	0 (0.0)	1 (1.0)
People in my neighborhood are willing to help their neighbors.				0.350
Strongly agree	23 (27.4)	22 (29.7)	1 (10.0)
Agree	46 (54.7)	40 (54.1)	6 (60.0)
Neither agree nor disagree	4 (4.8)	4 (5.4)	0 (0.0)
Disagree	5 (6.0)	4 (5.4)	1 (10.0)
Strongly disagree	2 (2.4)	1 (1.4)	1 (10.0)
Do not know	4 (4.8)	3 (4.1)	1 (10.0)

Data presented as n (%).

SEDOH = socioecological determinants of health.

Successful vs unsuccessful cecal intubation.

Having unsuccessful cecal intubation was associated with a lower employment rate (22.2% vs 37.3%; p = 0.035), lower physician trust (55.6% vs 73.3%; p = 0.040), and fewer yearly physician consultations (88.9% vs 100.0%; p = 0.004) compared to having a successful cecal intubation. In contrast, patients with successful cecal intubation were more confident in preventing health-related problems (53.3% vs 33.3%; p = 0.020) and had a more supportive social environment (72.0% vs 66.7%; p = 0.001) compared to those with unsuccessful cecal intubation (Table 3). Adequate bowel preparation quality was highly associated with successful cecal intubation (p < 0.01; data not shown).

TABLE 3.

Comparison of SEDOHs by successful vs unsuccessful cecal intubation (N = 84)

Variable	Overall (N = 84)	Successful cecal intubation (N = 75)	Unsuccessful cecal intubation (N = 9)	p
Employment status				0.035
Employed	30 (35.7)	28 (37.3)	2 (22.2)
Unemployed	1 (1.2)	1 (1.3)	0 (0.0)
Retired	30 (35.7)	28 (37.3)	2 (22.2)
Student	1 (1.2)	0 (0.0)	1 (11.1)
Disabled	19 (22.6)	15 (20.0)	4 (44.4)
Other	3 (3.6)	3 (4.0)	0 (0.0)
Do you have difficulty understanding written medical information?				0.340
Always	4 (4.8)	3 (4.0)	1 (11.1)
Often	3 (3.6)	2 (2.7)	1 (11.1)
Sometimes	9 (10.7)	9 (12.0)	0 (0.0)
Occasionally	8 (9.5)	8 (10.7)	0 (0.0)
Never	60 (71.4)	53 (70.6)	7 (77.8)
How often do you have someone help you read the hospital materials?				0.090
Always	9 (10.7)	7 (9.3)	2 (22.2)
Often	5 (6.0)	3 (4.0)	2 (22.2)
Sometimes	10 (11.9)	10 (13.3)	0 (0.0)
Occasionally	10 (11.9)	10 (13.3)	0 (0.0)
Never	50 (59.5)	45 (60.0)	5 (55.6)
I am the person responsible for taking care of my health.				0.650
Strongly disagree	1 (1.2)	1 (1.3)	0 (0.0)
Disagree	2 (2.4)	2 (2.7)	0 (0.0)
Agree	40 (47.6)	34 (45.3)	6 (66.7)
Strongly agree	41 (48.8)	38 (50.7)	3 (33.3)
I am confident that I can help prevent or reduce problems associated with my health.				0.020
Disagree	4 (4.8)	4 (5.3)	0 (0.0)
Agree	43 (51.2)	40 (53.3)	3 (33.3)
Strongly agree	36 (42.9)	31 (41.3)	5 (55.6)
NA	1 (1.2)	0 (0.0)	1 (11.1)
Do you have internet access?				0.120
Yes	71 (84.5)	65 (86.7)	6 (66.7)
No	13 (15.5)	10 (13.3)	3 (33.3)
If you needed it, how often is someone available to turn to for suggestions about how to deal with a personal problem?				0.080
None of the time	1 (1.2)	1 (1.3)	0 (0.0)
A little of the time	4 (4.8)	2 (2.7)	2 (22.2)
Some of the time	9 (10.7)	9 (12.0)	0 (0.0)
Most of the time	12 (14.3)	10 (13.3)	2 (22.2)
All of the time	58 (69.0)	53 (70.7)	5 (55.6)
If you needed it, how often is someone available who understands your problems?				0.870
A little of the time	5 (6.0)	5 (6.7)	0 (0.0)
Some of the time	8 (9.5)	7 (9.3)	1 (11.1)
Most of the time	15 (17.9)	13 (17.3)	2 (22.2)
All of the time	56 (66.7)	50 (66.7)	6 (66.7)
Your doctor only thinks about what is best for you.				0.043
Strongly disagree	1 (1.2)	1 (1.3)	0 (0.0)
Disagree	1 (1.2)	0 (0.0)	1 (11.1)
Neutral	7 (8.3)	5 (6.7)	2 (22.2)
Agree	60 (71.4)	55 (73.3)	5 (55.6)
Strongly agree	14 (16.7)	13 (17.3)	1 (11.1)
Unsure	1 (1.2)	1 (1.3)	0 (0.0)
Are you currently covered by any of these insurance plans?				0.002
Medicare	31 (36.9)	28 (37.3)	3 (33.3)
Medicaid	8 (9.5)	4 (5.3)	4 (44.4)
Blue Cross/Blue Shield of Alabama	28 (33.3)	26 (34.7)	2 (22.2)
Other private insurance	17 (20.2)	17 (22.7)	0 (0.0)
How easy is it for you to make an appointment if you are sick and need health care?				0.480
Very easy	50 (59.5)	43 (57.3)	7 (77.8)
Easy	32 (38.1)	30 (40.0)	2 (22.2)
Neither easy nor difficult	2 (2.4)	2 (2.7)	0 (0.0)
How long has it been since you last saw a health care professional about your health?				0.004
Within the past year	83 (98.8)	75 (100.0)	8 (88.9)
Never or unsure	1 (1.2)	0 (0.0)	1 (11.1)
People in my neighborhood are willing to help their neighbors.				0.830
Strongly agree	23 (27.4)	21 (28.0)	2 (22.2)
Agree	46 (54.7)	41 (54.7)	5 (55.6)
Neither agree nor disagree	4 (4.8)	4 (5.3)	0 (0.0)
Disagree	5 (6.0)	4 (5.3)	1 (11.1)
Strongly disagree	2 (2.4)	2 (2.7)	0 (0.0)
Do not know	4 (4.8)	3 (4.0)	1 (11.1)

Data presented as n (%).

SEDOH = socioecological determinants of health.

No adenoma vs adenoma.

Patients with no adenoma(s) detected reported better social support compared to those who had adenomas detected, as they were more likely to have someone available to talk to (80.0% vs 57.9%; p = 0.020), someone who understood their problems (77.8% vs 52.6%; p = 0.030), and had supportive neighbors in their community (64.4% vs 42.1%; p = 0.060; Table 4).

TABLE 4.

Comparison of no adenoma detected vs adenoma(s) detected (N = 84)

Variable	Overall (N = 84)	No adenoma detected (N = 45)	Adenoma(s) detected (N = 38)	p
Employment status				0.300
Employed	30 (35.7)	14 (31.1)	15 (39.5)
Unemployed	1 (1.2)	0 (0.0)	1 (2.6)
Retired	30 (35.7)	15 (33.3)	15 (39.5)
Student	1 (1.2)	1 (2.2)	0 (0.0)
Disabled	19 (22.6)	14 (31.1)	5 (13.2)
Other	3 (3.6)	1 (2.2)	2 (5.3)
Do you have difficulty understanding written medical information?				0.400
Always	4 (4.8)	4 (8.9)	0 (0.0)
Often	3 (3.6)	2 (4.4)	1 (2.6)
Sometimes	9 (10.7)	4 (8.9)	5 (13.2)
Occasionally	8 (9.5)	4 (8.9)	4 (10.5)
Never	60 (71.4)	31 (68.9)	28 (73.7)
How often do you have someone help you read the hospital materials?				0.090
Always	9 (10.7)	3 (6.7)	6 (15.8)
Often	5 (6.0)	4 (8.9)	1 (2.6)
Sometimes	10 (11.9)	8 (17.8)	2 (5.3)
Occasionally	10 (11.9)	3 (6.7)	7 (18.4)
Never	50 (59.5)	27 (60.0)	22 (57.9)
I am the person responsible for taking care of my health.				0.780
Strongly disagree	1 (1.2)	1 (2.2)	0 (0.0)
Disagree	2 (2.4)	1 (2.2)	1 (2.6)
Agree	40 (47.6)	22 (48.9)	17 (44.7)
Strongly agree	41 (48.8)	21 (46.7)	20 (52.6)
I am confident that I can help prevent or reduce problems associated with my health.				0.580
Disagree	4 (4.8)	2 (4.4)	2 (5.3)
Agree	43 (51.2)	25 (55.6)	17 (44.7)
Strongly agree	36 (42.9)	18 (40.0)	18 (47.4)
N/A	1 (1.2)	0 (0.0)	1 (2.6)
Do you have internet access?				0.560
Yes	71 (84.5)	37 (82.2)	33 (86.8)
No	13 (15.5)	8 (17.8)	5 (13.2)
If you needed it, how often is someone available to turn to for suggestions about how to deal with a personal problem?				0.022
None of the time	1 (1.2)	0 (0.0)	1 (2.6)
A little of the time	4 (4.8)	0 (0.0)	4 (10.5)
Some of the time	9 (10.7)	2 (4.4)	7 (18.4)
Most of the time	12 (14.3)	7 (15.6)	4 (10.5)
All of the time	58 (69.1)	36 (80.0)	22 (57.9)
If you needed it, how often is someone available who understands your problems?				0.031
A little of the time	5 (6.0)	3 (6.7)	2 (5.3)
Some of the time	8 (9.5)	1 (2.2)	7 (18.4)
Most of the time	15 (17.9)	6 (13.3)	9 (23.7)
All of the time	56 (66.7)	35 (77.8)	20 (52.6)
Your doctor only thinks about what is best for you.				0.590
Strongly disagree	1 (1.2)	1 (2.2)	0 (0.0)
Disagree	1 (1.2)	1 (2.2)	0 (0.0)
Neutral	7 (8.3)	2 (4.4)	4 (10.5)
Agree	60 (71.4)	33 (73.3)	27 (71.1)
Strongly agree	14 (16.7)	7 (15.6)	7 (18.4)
Unsure	1 (1.2)	1 (2.2)	0 (0.0)
Are you currently covered by any of these insurance plans?				0.720
Medicare	31 (36.9)	18 (40.0)	12 (31.6)
Medicaid	8 (9.5)	5 (11.1)	3 (7.9)
Blue Cross/Blue Shield of Alabama	28 (33.3)	13 (28.9)	15 (39.5)
Other private insurance	17 (20.2)	9 (20.0)	8 (21.0)
How easy is it for you to make an appointment if you are sick and need health care?				0.031
Very easy	50 (59.5)	21 (46.7)	28 (73.7)
Easy	32 (38.1)	22 (48.9)	10 (26.3)
Neither easy nor difficult	2 (2.4)	2 (4.4)	0 (0.0)
How long has it been since you last saw a health care professional about your health?				0.360
Within the past year	83 (98.8)	44 (97.8)	38 (100.0)
Never or unsure	1 (1.2)	1 (2.2)	0 (0.0)
People in my neighborhood are willing to help their neighbors.				0.060
Strongly agree	23 (27.4)	13 (28.9)	10 (26.3)
Agree	46 (54.8)	29 (64.4)	16 (42.1)
Neither agree nor disagree	4 (4.8)	1 (2.2)	3 (7.9)
Disagree	5 (6.0)	0 (0.0)	5 (13.2)
Strongly disagree	2 (2.4)	1 (2.2)	1 (2.6)
Do not know	4 (4.8)	1 (2.2)	3 (7.9)

Data presented as n (%).

DISCUSSION

In this retrospective observational study, we have demonstrated significant associations between SEDOHs and the quality metrics of colonoscopy among rural surgical populations in Alabama. Specifically, a lack of health literacy, insurance coverage, social support, and physician trust were more likely among patients with low-quality colonoscopy metrics. These findings underscore the novelty of the SEDOH-88 for assessing multidimensional factors that are not typically associated with colonoscopy quality metrics. Overall, assessing these SEDOHs and linking them to colonoscopy quality metrics helps inform strategies to improve colonoscopy quality in rural settings at the patient, provider, and institutional level.

From the patients’ perspective, bowel preparation is considered the most challenging part of undergoing a colonoscopy,²⁵ leading to a failure to follow preparation instructions, which is an independent predictor of low-quality bowel preparation.¹⁷ We demonstrated that patients with suboptimal bowel preparation were more likely to experience difficulty understanding written medical information and more often required assistance reading hospital materials compared to patients with optimal bowel preparation. Similarly, Nguyen and Wieland demonstrated that interpreter requirement (poor English language literacy) was significantly higher among those with inadequate bowel preparation (42.2% vs 7.1%; p < 0.0001) and conferred increased odds of low-quality preparation (OR 15.17, 95% CI, 4.49–20.6; p < 0.0001) in a study of 300 consecutive average-risk screening colonoscopies.¹⁷ As health information is increasingly being accessed via online platforms,²⁶ we noted a significant digital divide in our study as patients with suboptimal bowel preparation described limited internet access compared to those with optimal bowel preparation. Although this finding may reflect infrastructural disparities that need to be addressed at the system level, providers can leverage interventions such as patient navigators and focused education to assist with bowel preparation instructions and overcome limited health literacy, language, and internet access barriers.²⁷ Overall, these findings underscore the importance of health literacy-sensitive care among patients undergoing a colonoscopy. Additional examples of such health literacy-sensitive care include promoting communication strategies to avoid medical jargon, speaking at a cadence that is easy to follow, using illustrations and educational materials at the recommended sixth grade reading level, and teach-back techniques with patients and their family members.²⁸

Beyond health literacy, adherence to bowel preparation instructions is influenced by multiple other patient and provider-related factors ranging from self-efficacy, palatability of the bowel preparation, GI side effect profile, and volume of preparation.²⁹ These barriers to adequate bowel preparation may be addressed by measuring and targeting social support mechanisms. Prior studies have highlighted how personal relationships can be associated with a higher likelihood of adherence to bowel preparation instructions.¹⁶ We observed that patients with suboptimal bowel preparation had worse overall social environments, including a lower employment rate, higher likelihood of disability, and a diminished sense of health responsibility compared to those with optimal preparation. In contrast, those with successful cecal intubations were noted to be more confident in preventing health-related problems and had a more supportive social environment.

Having patient–physician trust has been reported to play an important role in improving the likelihood of undergoing CRC screening.³⁰ Although many factors influence a patient’s level of physician trust, such as racial concordance with their provider, routine clinical visits provide opportunities to cultivate physician trust through effective communication.³¹ In concert with these observations, we demonstrated that patients with unsuccessful cecal intubations had lower physician trust and fewer yearly physician visits compared to those with successful cecal intubations. Compared to the outcome of adequate quality of bowel preparation (88% of cohort), successful cecal intubation rates (89% of cohort) can be influenced both by quality of bowel preparation and physician-level factors, including clinical experience and technical skill.³² Although we are limited in our ability to compare colonoscopy quality across the limited providers in this small study, these findings highlight a potentially important interplay between physicians and patients that should be measured for physician feedback. Future multilevel interventions that focus on health communication strategies can offer mechanisms to enhance physician trust and downstream outcomes.³³

Higher ADR has been consistently associated with colonoscopy effectiveness in reducing CRC incidence and mortality.⁸ Despite 10.7% of patients having unsuccessful cecal intubations, we demonstrated an ADR of 45.8%. Although ADR is impacted by patient-related (eg, sex, age, race/ethnicity),³⁴ provider-related (eg, experience, medical specialty, fatigue),³⁵ and examination-related factors (eg, withdrawal time, timing of colonoscopy),³⁶ our findings build on prior reports noting social support and health care access to be influential.¹⁶ Patients who had adenoma(s) detected were less likely to have someone available who understood or to help address their personal problems. The association of social cohesiveness and adenoma detection may reflect an influence on prior preventive behaviors, decreasing these adenoma rates or potentially mitigating participation in high-risk behaviors such as smoking, alcohol consumption, and physical inactivity, which are known risk factors for the development of CRC.³⁷

Insurance coverage is known to influence multiple aspects of the colonoscopy process, including access to CRC screening and follow-up rates, particularly among individuals lacking supplemental coverage.³⁸ In concert with these observations and the 2003 Institute of Medicine’s report recognizing the pervasive effects of inadequate insurance coverage,³⁹ we demonstrated that patients with suboptimal bowel preparation were more likely to be insured with Medicaid compared to those with optimal bowel preparation. Although this observation may be influenced by the lower rates of employment and lack of social support among this group, previous studies have demonstrated that Medicaid independently confers increased odds of suboptimal bowel preparation.¹⁶ The link between Medicaid insurance and the disparity in bowel preparation quality is especially notable as Alabama is one of the few states that has not expanded Medicaid access following the Affordable Care Act,⁴⁰ which others have shown to improve access to CRC screening.⁴¹ Overall, the link between low-quality colonoscopy and Medicaid insurance status reflects that financial coverage is insufficient for promoting successful high-quality colonoscopy efforts. Successful programs such as state-sponsored CRC screening programs must also incorporate navigation components to address disparities in CRC mortality.⁴²

Our study has several limitations. First, we cannot infer causality between the identified SEDOHs and the low-quality colonoscopy metrics due to the retrospective, observational nature. As the time between colonoscopy completion and the date of survey administration varied among our cohort, we recognized the potential for recall bias. Although the response rate of 54.2% suggests a potential for response bias, 99.5% of all survey questions were completed. This finding aligns with the 80% minimum completion rate recommended by the American Association for Public Opinion Research,⁴³ and therefore, we believe this is an acceptable response rate for this socially vulnerable population. Second, the study may be underpowered to detect several meaningful associations between SEDOHs and colonoscopy metrics. The small sample size limits our ability to perform multivariable analysis to control for factors known to influence colonoscopy quality, such as age and sex. We are unable to comment on which SEDOH has the highest independent association with colonoscopy quality without multivariable analysis; however, our findings highlight novel associations that are important for further study in this understudied population. Third, we chose to include both screening and diagnostic colonoscopy procedures that may influence colonoscopy quality, and variables such as prior colonoscopy history were not available. Finally, because our cohort comprised only rural, English-speaking patients from Alabama, the results may not be generalizable to those residing in urban and suburban communities, non–English-speaking patients, and those residing in other US states with different social and economic microclimates.

CONCLUSIONS

Our study demonstrates that low health literacy, a lack of internet access, underinsured status, and a lack of physician trust were associated with low-quality colonoscopy, whereas having a supportive social environment and a higher patient sense of responsibility was associated with higher-quality colonoscopy metrics. By identifying the relationship between SEDOHs and colonoscopy quality metrics using the SEDOH-88, important targets exist for guiding quality improvement efforts to address disparities among vulnerable rural populations.

These findings were originally presented at the annual meeting of the American Society of Colon and Rectal Surgeons.⁴⁴