Videos Improve Patient Understanding of Chemotherapy Terminology in a Rural Setting

Abstract

Background:

It is critical patients understand the terms used to describe oncology treatments; however, even basic chemotherapy terminology can be misunderstood. Rural communities tend to have especially low levels of health literacy compared to non-rural communities. In order to address low health literacy in rural communities, this study tested rural participants’ understanding of previously developed educational chemotherapy videos that were designed for an underserved urban population. Participants were also asked for feedback to determine if the videos could be improved.

Methods:

50 English-speaking patients who reside in counties classified as rural according to the Rural-Urban Continuum Code designations (RUCC 4–9) participated in the study. Participants were asked to define six chemotherapy terms before and after viewing a short, animated video explaining the term in English. Rates of correct and incorrect definitions provided by participants were also compared to our previously published results from an urban cohort.

Results:

All participants had statistically significantly higher rates of correct definitions for all six terms following the video intervention. Palliative Chemotherapy understanding improved the most (10% correct pre-video and 76% post-video intervention). For each video, the majority of participants (77%−92%) suggested no changes to the videos.

Conclusion:

Given the prevalence of low health literacy in rural communities, it is important to have effective educational interventions to improve understanding of basic oncology treatment terminology. Our study demonstrates that short, educational videos, originally designed for an underserved urban population, can significantly improve understanding of commonly misunderstood chemotherapy terminology in a rural setting as well.

Lay Summary:

Chemotherapy terminology can be confusing for patients. Understanding can be especially difficult in areas with low health literacy, such as underserved urban and rural communities. To address this concern, we previously developed short, animated videos describing basic chemotherapy terminology, which improved patient understanding in an underserved urban setting. In this study, we tested the videos in a rural population and established their effectiveness. Participants in the rural setting were significantly more likely to correctly define all six tested terms after watching the videos. Educational tools for high-need populations are essential to ensure patients can understand the treatment they receive.

Precis for table of contents:

Short animated videos, which improved patient understanding of chemotherapy terminology in an underserved urban setting, were also successful in significantly improving understanding in a rural setting.

Introduction:

Patient health literacy is important for informed consent¹, treatment adherence,^2,3 and improved clinical outcomes in oncology⁴. However, patients undergoing cancer treatment frequently misunderstand basic cancer terminology, including important information about their treatment plans and options⁵. Given that chemotherapy continues to be a widely used cancer treatment, improving patient understanding of common chemotherapy terms is critical in promoting ethical patient care¹ and optimizing outcomes⁴.

In the United States, 36% of adults have below basic or basic levels of health literacy⁶, with a high concentration of low health literacy in rural areas⁷. These low rates of health literacy can further perpetuate the disproportionally worse cancer outcomes that rural areas tend to have compared to non-rural areas^8–10. Given that approximately 14% of the United States population resides in rural counties according to the RUCC criteria^8,11 and that low health literacy is correlated with worse cancer outcomes¹², it is imperative that efforts are made to improve health literacy in these rural communities.

As of 2019, 17% of the population in Georgia reside in rural counties¹³, which is higher than the national average^8,11. To address the issue of health literacy in rural communities, and given the sizeable rural population in Georgia, we designed a rural population study to test our chemotherapy-related educational videos, which were previously designed for and were proven to significantly improve patient understanding of chemotherapy terminology in an underserved urban hospital¹⁴.

In our previous study, we interviewed 50 patients and using Kilbridge et al.’s methodology¹⁵, we developed a word bank of difficult to understand chemotherapy terms. We also asked 15 providers to identify terms that are important to understand. Fifty terms were identified by two or more patients or providers. An expert panel, including health literacy experts, decided to reduce the 50 to 20 terms for further testing. They based their choices on frequency of mention and term importance. The 20 terms were then tested with 50 patients confirming high levels of patient misunderstanding¹⁴. For example, 74% of patients did not understand ‘cancer’. We then developed, one-minute long videos in English to explain each of the 20 terms. To decrease participant burden, only six of these videos- Palliative Chemotherapy, Curative Treatment, Cancer, Blood Count, Risk of Infection, and Chemotherapy- were selected by the expert panel to be used to assess pre- and post-video intervention understanding. The videos significantly improved patient understanding of all six terms.

Our previous study contributed to the substantial body of literature that captures the efficacy of video-based educational interventions in oncology. Videos have been shown to improve cancer knowledge^16–20, screening and prevention behaviors^20–23, pain management²⁴, treatment adherence^24,25, and other clinical outcomes^18,21. Educational videos can be especially useful in low health literacy or undereducated populations^19,26 and have been well-received by patients²⁶.

Rural and underserved urban populations usually have low levels of health literacy^7,14,27,28, but different demographic compositions⁷, so we thought it important to test the videos in a rural setting. We hypothesized that the increase in understanding of the terms post-video compared to pre-video would be similar to that found in our previous study.

Materials and Methods:

We defined ‘rural’ counties according to the United States Department of Agriculture’s 2013 Rural-Urban Continuum Codes (RUCC)²⁹, which is their most recent classification system. This system distinguishes metropolitan counties and non-metropolitan counties by population size, degree of urbanization, and proximity to an urban area²⁹ (Table 1). The classification system contains 9 parts, with 3 metropolitan designations (1–3) and 6 non-metro categories (4–9). We included all 6 non-metro categories in our rural designation. While several methods for rural classification have been documented for health research, the RUCC system remains a standard classification system^30,31, and studies have shown high levels of agreement between the RUCC system and other measures³⁰.

Table 1.

Rural-Urban Continuum Code Designations²⁹

RUCC	Designation	Description
1	Metro	Counties in metro areas of 1 million population or more
2	Metro	Counties in metro areas of 250,000 to 1 million population
3	Metro	Counties in metro areas of fewer than 250,000 population
4	Non-metro	Urban population of 20,000 or more, adjacent to a metro area
5	Non-metro	Urban population of 20,000 or more, not adjacent to a metro area
6	Non-metro	Urban population of 2,500 to 19,999, adjacent to a metro area
7	Non-metro	Urban population of 2,500 to 19,999, not adjacent to a metro area
8	Non-metro	Completely rural or less than 2,500 urban population, adjacent to a metro area
9	Non-metro	Completely rural or less than 2,500 urban population, not adjacent to a metro area

Three clinics in different locations in Georgia that are part of the Winship Network³² agreed to participate in the study. An ethics researcher visited the infusion center of each to recruit patients. Each potential participant was asked to identify the county they lived in; eligible patients were those from a county that had a RUCC of 4–9. Verbal consent was obtained for the interview, which was then conducted during their infusion visit, as we have done in previous studies^14,33. To ensure privacy, the researchers positioned themselves near the patient and shared the video on an iPad that the patient could hold. The same six videos we tested in our previous study were tested in this current study. Before viewing the videos, participants were asked if they knew the term that was described in the video. Unknown words were scored incorrect. If the term was known, the participants were asked to define the term in their own words. Participants were then shown the video and asked to define the term again. Finally, participants were shown the video a second time to solicit suggested changes to the videos. All interviews were audio-recorded. The researcher scored the answers during the interview as correct or incorrect using the video definition of the term. A second researcher then listened to the audio-recording and independently coded the participants’ answers as correct or incorrect. Twelve discrepancies were resolved by a third researcher. The sections of the audio-recordings in which the participants suggested changes to the videos were transcribed verbatim. Reviewing the transcripts, two investigators independently generated inductive codes capturing the themes of the suggested changes. Discrepancies in the codes were discussed and a final codebook was made. All transcripts were then independently coded and double coded by two researchers using the code book. All 10 discrepancies were resolved by consensus.

Participants also completed the Rapid Estimate of Adult Literacy in Medicine (REALM) scale to assess their health literacy. The REALM assessment presents participants with 66 common medical terms, with literacy determined by the number correctly pronounced (Table 2)³⁴. REALM scores were analyzed as both continuous and categorical variables.

Table 2.

Summary of participant demographics and comparison to previous urban cohort

Demographic	Rural sites n (%)	Urban hospital n (%)	P-value
Mean age	63	54	<0.001
RUCCa
4–5	19 (38)	–
6–8	31 (62)	–
Sex			0.108
Female	25 (52)	34 (68)
Male	23 (48)	16 (32)
Race/Ethnicity			<0.001
Black	10 (21)	47 (96)
White	38 (79)	2 (4)
Missing	2
Education			0.246
High school or less	28 (56)	33 (67)
Some college	22 (44)	16 (33)
Income			0.001
≤40,000	24 (59)	40 (89)
>40,000	17 (42)	5 (11)
Missing	9
Employment			0.014
Employed	10 (20)	2 (4)
Retired/unemployed/disabled	39 (80)	47 (96)
Missing	1
REALMb score (number of words pronounced correctly)			<0.001
<3rd grade level (0–18)	1 (2)	2 (4)
4th–6th grade level (19–44)	0 (0)	8 (17)
7th–8th grade level (45–60)	7 (16)	19 (40)
High school level (61–66)	35 (81)	18 (38)
REALM score average	62	53	0.001

a. Abbreviation:

^aRUCC, Rural-Urban Continuum Codes.

^bREALM, Rapid Estimate of Adult Literacy in Medicine.

b. The p-value is calculated by ANOVA for numerical covariates, and chi-square test or Fisher’s exact test for categorical covariates, where appropriate.

Rates of understanding of the six definitions are reported before and after the video, including 95% confidence intervals for the rates. The pre- and post-video rates of understanding were compared using McNemar’s test. Patient characteristics were compared across hospital setting and post-video understanding of terms using ANOVA, chi-square tests, or Fisher’s exact tests, where appropriate. Significance was assessed at the 0.05 level, and all tests were be two-sided. Statistical analysis was performed using SAS 9.4(SAS Institute Inc., Cary, NC). Power and sample size: To detect that a paired difference of 30% between the pre-test and post-test for the most understood word, chemotherapy (42% vs. 72%), with 80% power, a Type I error of 0.05, and an assumed percent discordant of 54%, we required at least 50 patients. The estimate for percent discordant, the proportion of pairs in which the response differed, is based on a conservative assumption of independence between tests.

Results:

This study was approved by the institution’s Institutional Review Board (IRB00085515). Of the 142 patients approached, 21 did not reside in a rural county and 7 had not consented for chemotherapy and were therefore ineligible. Fifty (50/114, 44%) of the eligible patients agreed to participate. The most common reasons for refusals were not being interested in participating (19/64, 30%), or not having enough time (15/64, 23%). Other reasons included being too tired (10/64, 16%), and having trouble hearing or seeing the videos due to vision or hearing impairments (5/64, 8%). A summary of the participant demographic information and a comparison to our original cohort of urban participants are reported in Table 2.

All participants had significantly higher rates of correct definitions for all six terms after watching the videos (Table 3). Participants had the most significant increase in correctly defining ‘palliative chemotherapy’ in the post-video versus the pre-video assessment, with only 10% of participants correctly defining the term before the video and 76% correctly defining the term post-video(Table 3).

Table 3.

Percentages of rural participants who provided correct definitions pre-and post-video intervention

Term	Before – Rate (95% CI)	After – Rate (95% CI)	McNemar p-value
Cancer	0.32 (0.20 – 0.47)	0.76 (0.62 – 0.87)	<0.001
Chemotherapy	0.52 (0.37 – 0.66)	0.84 (0.71 – 0.93)	<0.001
Palliative chemotherapy	0. 10 (0.03 – 0.22)	0.76 (0.62 – 0.87)	<0.001
Curative treatment	0.35 (0.22 – 0.50)	0.76 (0.62 – 0.87)	<0.001
Blood count	0.55 (0.40 – 0.69)	0.90 (0.78 – 0.97)	<0.001
Risk of infection	0.54 (0.39 – 0.68)	0.82 (0.69 – 0.91)	<0.001

Abbreviation: CI, Confidence Interval

Before the video intervention, participants with an income greater than $40,000 per year were significantly more likely to correctly define ‘risk of infection’ (76% vs. 38%, p=0.014) and ‘curative treatment’ (69% vs. 13%, p<0.001). Participants who had at least some college education were significantly more likely to correctly define ‘cancer’ (50% vs. 18%, p=0.016) and ‘curative treatment’ (52% vs. 21%, p=0.024) in the pre-video assessment. White participants were significantly more likely to correctly define the ‘risk of infection’ term in the pre-video assessment compared to Black participants (p=0.029).

Significant associations between rates of correct definitions given by participants in the post-video assessment and demographic measures are reported in Table 4. Of all the demographics, REALM score of high-school readability or higher (average REALM score of at least 61) was the most frequently associated with participants correctly defining terms post-video. Only ‘Curative Treatment’ did not yield a significant association between higher REALM scores and a correct definition (Table 4).

Table 4.

Significant associations between key demographics and participant correct definitions post-video intervention

Post-video Correct Understanding n (%)
Demographic	Chemo-therapy	P Value	Risk of infection	P value	Blood count	P value	Cancer	P value	Curative treatment	P value	Palliative chemotherapy	P value
Sex
Female	21 (84)	1.00	24 (96)	0.009	25 (100)	0.017	18 (72)	0.617	17 (68)	0.243	21 (84)	0.133
Male	19 (83)		15 (65)		17 (77)		18 (78)		19 (83)		15 (65)
Education
<High school	21 (75)	0.064	19 (68)	0.003	23 (85)	0.362	18 (64)	0.029	17 (61)	0.004	17 (61)	0.004
Some college	21 (95)		22 (100)		21 (95)		20 (91)		21 (95)		21 (95)
Income
≤40,000	19 (79)	0.373	17 (71)	0.029	21 (88)	1.00	15 (63)	0.028	16 (67)	0.152	15 (63)	0.005
>40,000	16 (94)		17 (100)		15 (88)		16 (94)		15 (88)		17 (100)
REALMa score mean
Correct	63	0.009	63	0.003	63	<.001	63	0.031	63	0.100	63	0.006
Incorrect	55		54		50		57		58		56

a. Abbreviation:

^aREALM, Rapid Estimate of Adult Literacy in Medicine

b.The p-value is calculated by ANOVA for numerical covariates, and chi-square test or Fisher’s exact test for categorical covariates, where appropriate.

We also compared the rates of correct definitions of this rural cohort and our previous urban cohort. In the pre-video assessment, rural participants were more likely to correctly define ‘risk of infection’ (54% vs. 32%, p=0.026), ‘blood count’ (55% vs. 22%, p <0.001), and ‘cancer’ (32% vs. 8%, p=0.003). In the post-video assessment, the only significant correlation seen was for the blood count video, where rural participants were more likely to correctly define ‘blood count’ after viewing the video than urban participants (90% vs. 74%, p=0.042).

Forty-eight of the fifty participants were asked for feedback on the videos. 29 participants (60%) gave feedback on one or more of the videos, for a total of 53 comments. Twenty-seven(51%) of those suggestions were to elaborate on information in the video, such as stating in the Risk of Infection video that a cancer patient with a high fever needs to go to the hospital, or in the Chemotherapy video that chemotherapy can be given through a port. Other suggestions included adjusting animation features, such as adding a face to the human pictured in the Blood Count video (7 mentions, 13%), adding a definition, such as defining neutropenia in the Risk of Infection video (6 mentions, 11%), changing a word (3 mentions, 6%), such as changing the word ‘kill’ in the phrase ‘chemotherapy can kill other rapidly dividing cells’ to something less strong. Six suggestions (11%) stated that information in the video may be inaccurate. Each suggestion was reviewed by a science expert and the video version was deemed correct.

Five comments (9%) were made about the tone of the video. Three of these comments were made about the Palliative Chemotherapy video: one participant suggested making the tone more optimistic, another participant suggested that the patient featured in the video should be less excited about a treatment that would not cure their cancer, and the final participant noted the tone of the video may be too simplistic for some adults. One participant indicated that the chemotherapy video should be more optimistic about the new targeted approaches of chemotherapy, although it was determined by the coders that the patient was referring to immunotherapy, not a novel chemotherapy. Finally, one patient mentioned that the Risk of Infection video should not urge patients to avoid stress during treatment, but instead the video should acknowledge that stress is inevitable for cancer patients.

While 29 participants provided useful feedback for the six videos, the majority of participants gave no feedback: 37(77%) participants gave no feedback on the Chemotherapy video, 44(92%) participants gave no feedback on the Curative Treatment video, 40 (79%) participants gave no feedback on the Risk of Infection, Cancer, and Palliative Chemotherapy videos, and 40 (83%) participants gave no feedback on the Blood Count video.

The videos can be found at https://www.cancerquest.org/media-center/videos/cancer-treatment-terms.

Discussion:

The goal of this study was to evaluate the efficacy of short, simple videos in improving understanding of commonly misunderstood chemotherapy terminology in rural Georgia and to query the rural participants about suggested changes to the videos. The results are consistent with the extensive literature on the efficacy of educational videos^{16,18,19,21,23–26} and our hypothesis that the videos would significantly improve participant understanding of all six chemotherapy terms as they had in our previously published study¹⁴.

We established in our previous study that most patients incorrectly defined the six terms before the video intervention in an urban setting¹⁴. Rates of correctly defining the term before the video were higher in the rural setting for three of the videos–Risk of Infection (p=0.026), Blood Count(p<0.001), and Cancer(p=0.003)–but comparable in the pre-video assessment between the urban and rural cohorts for the other three videos. While rates of correct definitions in the pre-video assessment were higher in the rural cohort for those three terms, only about half of participants in the rural cohort correctly defined those three terms and only 5 participants (10%) correctly defined ‘palliative chemotherapy’, demonstrating that educational interventions are still necessary to improve understanding.

The notable variations between the rural and urban cohorts in the rates of correct definitions pre-video may be due to socioeconomic differences^7,28. This rural cohort of patients had significantly higher income (p=0.001) and were more likely to be employed (p=0.014) compared to the urban cohort(Table 2). These results are consistent with findings from other studies which established that higher socioeconomic status is associated with higher rates of health literacy^7,28, which was borne out by the higher REALM scores in the rural cohort(p= 0.001).

The higher REALM for our rural participants may have been due to patients with lower health literacy having lower interest in and lower participation in clinical research^35,36, even after adjusting for sociodemographic factors³⁵. However, other research suggests that willingness to participate in interview studies compared to more invasive research is higher for individuals with low health literacy, especially if the research is conducted by their treatment hospital³⁷. Since this was our first study with Winship Network sites, we may have been perceived more as visitors than we were at the urban site where we have conducted multiple studies, which may have resulted in lower literate individuals not participating. Since we could not test refusers’ literacy, we cannot determine if this was in fact the case. In spite of these differences in demographics, which we anticipated and were in part the motivation for this study⁷, the videos significantly improved understanding of all six terms for the rural cohort, demonstrating their usefulness in this rural setting.

We also queried participants for feedback on the videos to determine if they needed to be improved. For each video, the majority of participants had no comments; however, 53 helpful comments were received, which were carefully reviewed. We plan to make all the suggested changes in the next iteration of the videos.

A few notable limitations to this study exist. First, only six of the twenty developed videos were tested. Additionally, the study sample was limited to participants who were English-speaking. To ensure these videos are more widely available to patients, we are applying for a grant to translate the videos into Spanish, as our first translation, given that 9.9% of Georgia’s residents are Hispanic³⁸. Finally, findings are not necessarily generalizable to other rural areas due to demographic variations between different rural communities⁷. However, since these videos were validated in two cohorts of patients with different demographics and socioeconomic profiles, the potential for their widespread efficacy is promising. We look forward to testing these videos in additional populations, such as pediatric oncology.

In conclusion, our study validated that these short educational videos improve understanding of commonly misunderstood chemotherapy terminology in a rural setting. Our results are consistent with our previously published study and others that show short, animated videos are a viable tool to improve chemotherapy health literacy.