Publication/Presentation Date

7-25-2014

Abstract

Abstract

Clinical research trials may benefit from attracting a sample population that accurately reflects the total population. Over 600 paper surveys have been administered to patients across several practices in the Lehigh Valley Health Network (LVHN). The surveys provide feedback on patients’ past experiences and personal views towards clinical trials. They also include demographic information to show trends in the data. Another set of surveys will also be sent to research professionals across the nation in federally funded institutions. These surveys will provide insight on researchers’ tendencies and beliefs when enrolling patients in clinical trials. The study will be open until 2015, but the data that has come out of the patient survey so far is not showing any significant difference of opinions towards participation in clinical research between English and non-English speaking patients nor between college educated and non-college educated patients.

Factors Influencing Participation in Clinical Research

Clinical research trials are a crucial piece of the healthcare field. In order to ensure productive trials for new drugs and other programs, the pilot studies should enroll patients who demographically represent the population that will ultimately benefit from the product. Recent studies have shown that the rates in which minority groups and women participate in research are much lower than those of Caucasians and males, respectively. This partially stems from the Food and Drug Administration’s 1977 decision to prohibit the participation of fertile and pregnant women in Phase I drug trials. However, this demographic is important in clinical trials; they may respond to drugs differently, and this information is best obtained in the pilot stages of a medication or apparatus. This study aims to discover the attitudes of patients towards participation in clinical research. This will allow researchers adjust their methods accordingly, and ultimately, this can help diversify research participants.

Literature Review

As mentioned, women and minorities have a low turnout in clinical research trials. Women represented less than 40% of participants in cancer research, according to a review1 conducted by Jagsi, Motomura, Amarnath, Jankovic, Sheets, and Ubel. Non-Caucasian women participate in less research than Caucasian women and have reported in higher numbers that they feel that doctors do not care about them2. Studies have also indicated that non-Caucasian men similarly participate in fewer research trials, as well.

Not only do more males tend to participate in clinical research as patients, but also men tend to conduct research more frequently. A survey3 conducted by Lloyd, Phillips, and Aber discovered that out of a class of 428 graduates from Penn State College of Medicine, only 41 females had conducted research. Another study4, conducted by Waisbre, Bowles, Hasan, Zou, Emans, Goldberg, and Christou, showed that only 26% of researchers submitting grant requests were women.

Methods

This study is a multi-faceted investigation into the views of both patients and researchers towards clinical research. The data is being collected in the form of two surveys: one for patients, and one for research professionals. The survey for research professionals has not yet begun, and it will not be discussed in depth.

Patient Surveys

Paper surveys have been administered to patients in ten different Lehigh Valley Health Network facilities, focusing on three major divisions: Family Medicine, Women’s Health, and Emergency Medicine. The research team was split up between Women’s Health and Family Medicine, and emergency room surveys are beginning to be collected now. This paper focuses on data from Women’s Health clinics. The surveys were translated into English, Spanish, and Traditional and Simplified Chinese in order to ensure minimal exclusion of demographics. A copy of the English survey is attached in Appendix A. By the end of the study, at least 400 completed surveys will be collected from each of the three divisions mentioned above for a statistically strong sample size. Selection criteria for participants simply dictate that they must be a patient of the clinic, and they must be eighteen or older.

Validity and Reliability

Pilot versions of both surveys were sent to thirty employees of the Lehigh Valley Health Network (fifteen for each survey) before the study began in full. To test the validity of the survey, a questionnaire was attached to the pilot surveys, in which participants were asked to write what they thought the purpose of the survey was. This ensured that survey was in fact addressing what it is supposed to address. To test the reliability, the same pilot survey was sent to the same thirty people in two waves, the second time being two weeks after the first. Pilot participants would complete the survey a second time, and the given responses would be cross-checked with their responses the first time they completed the survey. This tested to make sure participants did not change their answers significantly, ensuring survey reliability.

Results

Although an extensive data analysis will be done upon completion of the study, only simple, small-scale analysis has been done on the data at this point. All data comes only from Women’s Health clinics, which include the Maternal Fetal Medicine clinic (3900 Hamilton Boulevard, Suite 201, Allentown, PA 18103) and the Center for Women’s Medicine (17th and Chew, PO Box 7017, Allentown, PA 18105). Responses in questions J-1 (Relationship with Doctor), J-2 (Doctor’s Reputation in Community), K-1 (Distrust in Doctors), and K-2 (Time Commitment) have been compared between two demographic groups on two different axes: Language spoken at home (English vs. non-English) and Education (high school graduate or less vs. college or more). The question codes (J-1, J-2, etc.) are displayed in Table 3 for easy reference. On a Likert scale of 0 (least influence) to 4 (most influence), the mean rating for J-1 for the 392 English speakers surveyed was 3.22 with standard deviation 1.18. For the 322 non-English speakers, J-1 was 3.21 with standard dev. 1.26. Question J-2 received a score of 3.45 with standard dev. 1.06 for English speakers and 3.40 with standard dev. 1.14 for non-English speakers. Question K-1 received a score of 2.28 with standard dev. 1.59 for English speakers and 2.21 with standard dev. 1.58 for non-English speakers. Question K-2 received a 2.70 with standard dev. 1.26 for English speakers and 2.64 with standard dev. 1.25 for non-English speakers. This information can be viewed in Table 1 and Figure 1.

The same comparisons were made between those who reported having a high school diploma or GED, or less, as their highest degree of education, and those who reported having at least completed some college, or more. The results are as follows. J-1 scored 3.20 with standard deviation 1.24 for non-college participants and 3.21 with standard dev. 1.19 for college participants. J-2 scored 3.38 with standard dev. 1.13 for non-college participants and 3.44 with standard dev. 1.07 for college participants. K-1 scored 2.23 with standard dev. 1.59 for non-college participants and 2.28 with standard dev. 1.59 for college participants. K-2 scored 2.67 with standard dev. 1.30 for non-college participants and 2.71 with standard dev. 1.26 for college participants. These numbers are displayed in Table 2 and Figure 2.

No analysis has been done on the data from the research professionals’ survey, as this survey has not yet been sent. The study will remain open through 2015.

Conclusions

For the four questions analyzed, results displayed no significant difference between either group on either of the two axes. In other words, responses between English and non-English speaking patients were not significantly different, nor were responses between college educated and non-college educated patients. This means that patients somewhat universally feel the same way towards these four criteria in terms of motivation for participating in clinical research. In terms of the goal of the study, the data from these questions will not mandate any change in the procedures of clinical research, as these factors do not seem to play a role in excluding non-English speakers. It is important to note that this data comes from only women’s health clinics, so no results from any male participants were included in the data. Therefore, no conclusions can be drawn regarding a distinction in women’s perceptions versus men’s perceptions of these factors.

Limitations

Several limitations exist in this study. Most notably, for the patient survey, many patients have declined to complete the survey. This damages the data, as the people that will not complete the survey are also those that do not participate in clinical research. Of course, data from any patient is helpful, as the results have shown that beliefs amongst all patients tend to be similar (at least for certain questions, i.e. J-1, J-2, K-1, and K-2), but the study would benefit from learning the beliefs of unwilling patients as well. This limitation is difficult to circumvent, however, because coercing patients to take the survey could skew the results.

Another limitation is that due to the nature of surveying, some patients have left some portions of the survey incomplete. Accordingly, these incomplete surveys have been left out of the main pool of data. The data that patients do supply in their incomplete surveys can be useful, but if a patient leaves certain information, such as demographics, blank, there is no way to tie their responses to anything. By collecting at least 400 completed surveys from each practice, this should not be a major problem.

Tables and Figures

Table 1. Comparison of English and non-English Speaking Patients' Responses for Various Questions

Question

English Patient Mean Response

Non-English Patient Mean Response

J-1

3.22

3.21

J-2

3.45

3.40

K-1

2.28

2.21

K-2

2.70

2.64

Table 2. Comparison of College-Educated and Non-College Educated Patients' Responses for Various Questions

Question

College-Educated Patient Mean Response

Non-College Educated Patient Mean Response

J-1

3.21

3.20

J-2

3.44

3.38

K-1

2.28

2.23

K-2

2.71

2.67

Table 3. Questions Being Analyzed

Code

Question

J-1

Relationship with Doctor

J-2

Doctor's Reputation in Community

K-1

Distrust in Doctors

K-2

Time Commitment

Comments

Mentor: Anita Kurt

Department(s)

Research Scholars, Research Scholars - Posters

Document Type

Poster

Download

Share

COinS