Community Participation Report

Jump to another page... Executive Summary Hearing Community Voices Priority Rankings by Category Survey of Health Literacy in Bexar County San Antonio Health Literacy Initiative—Final Report Report on Community Meetings on Obesity Prevention Community Assessment for the Misión Evangélica Luterana Area ZIP Code 78227 Community Assessment Neighborhood Meetings Appendices Glossary

Survey of Health Literacy in Bexar County San Antonio Health Literacy Initiative—Final Report

Methods

The Short Test of Functional Health Literacy in Adults (STOFHLA) was used to measure health literacy. The STOFHLA uses a modified cloze procedure to literacy testing and measures subjects’ abilities in reading comprehension. The instrument requires that respondents read parts of an informed consent document and parts of an instruction sheet for a medical procedure. Approximately every 5th to 7th word is deleted in the passage and subjects are given a list of words to select from that best fit in the blank.

It was necessary to conduct in-person interviews because the health literacy test can only be administered in person. Two screening questions were completed prior to the survey to ensure that the respondent was an adult (18 years or older) and a resident of Bexar County. If the respondent answered “yes” to both questions and provided consent, s/he was administered the survey. The survey consisted of demographic items (age, marital status, education level, employment status, insurance status, etc.) and a short five-item scale to measure self-efficacy in relation to health literacy. Further, a single item language use question was included. The entire survey took no more than 10 to 12 minutes to administer.

All survey instruments were available in both English and Spanish, and the surveys were administered by trained adult volunteers from organizations involved with the SAHLI. All data collectors were required to attend a training session on survey administration. Additionally, the training sessions were conducted either by the investigators or the project coordinator and consisted of basic procedures for obtaining informed consent, interviewing respondents, and administering the STOFHLA. Depending on the experience of the interviewer, the training lasted from 15 to 30 minutes.

Cost considerations prevented conducting a door-to-door cluster sample to collect information about health literacy. When alternative data collection methods were being considered, it was estimated that a very high proportion of adults visit grocery stores at regular intervals. Therefore, it was determined that a convenience sample at grocery stores selected in a geographic distribution throughout the county could approximate a random sample. Permission to conduct the surveys was obtained from a national grocery/discount chain and from one local grocer. We conducted surveys at 11 of the national grocery/discount stores, and one local grocery store during a two week time period in November 2006. The stores were selected based on location to provide the broadest distribution of Bexar County residents possible. The map above shows the store locations.

Figure 1: Map of Data Collection Locations

Data collectors stood outside the stores and asked people to participate as they entered or exited the store. Efforts were made to have bilingual interviewers available at each location. At scheduled times, interviewers went to the various stores (e.g., mornings, afternoons, evenings, and weekends) to administer interviews. Respondents were given a $5 gift card to a local store for their participation. After 80% of the targeted number of surveys were completed, an analysis of respondent ZIP Codes was conducted to determine whether all areas of Bexar County were adequately represented. The result showed deficiencies in the southwest and southeast areas of the county. Canvassing local businesses in those areas helped with collection of additional interviews.

The Institutional Review Board at The University of Texas Health Science Center reviewed and approved this study.

Data Analysis

A total of 831 surveys were collected and data from these 831 surveys were entered into SPSS for review and analysis. Five surveys completed by minors and 40 surveys completed by residents with ZIP Codes outside Bexar County were excluded from the study, leaving a total of 786 surveys included in the study dataset. Surveys from ZIP Codes that overlap the Bexar County line were kept in the dataset. Figure 2 below illustrates survey distribution of sample respondents by ZIP Code and store location.

Figure 1: Map of Survey Distribution by ZIP Code

Data were reviewed and corrected against source documents for out-of-range data. Logically related variables such as text associated with other education, other employment, and other ethnicity were also checked and corrected. Several variables were recoded for the purposes of analysis. Raw Ethnicity data was recoded into a bivariate variable. A code of 1 indicates Hispanic and a code of 2 indicates Non-Hispanic. Additionally, language usage was recoded by excluding all non-Hispanic participants. STOFHLA scores were determined by adding the number of responses the participant answered correctly. Scores can range from 0 to 36, with higher scores indicating higher health literacy. Raw STOFHLA scores were distilled into a categorical variable using the following scale: scores from 0–16 were recoded as Inadequate, from 17–22 as Marginal, and from 23–36 as Adequate. A self efficacy score was calculated by taking the average of average scores of individual components of the self efficacy scale:

1. find health information, and understand the written health information;
2. understand the needed health information;
3. know who to ask for health information; and,
4. ask the doctor to explain health information.

Finally, education was recoded categorically into less than high school, graduate high school, and more than high school.

Results

Of the 786 surveys collected, 64% were completed by women. Overall, 50% of the participants were Hispanic, of which 90% were Mexican-American. Among Hispanics, 94% reported speaking predominantly English. The ages of the respondents ranged from 18 to 92 years with an average age of 42. Fifty-five percent of the respondents reported being married or living with someone. The total number of persons per household ranged from 1 to 20 with an average of 3.4. Table 1, below, summarizes these results.

Eighty-five percent of participants reported having at least a high school education, and 25% of these people reported having a college degree or doing post-graduate work. Additionally, 44% of the participants worked full time, and 33% reported household incomes over $40,000 per year. Seventy-two percent indicated that they currently have health insurance, and 77% reported that their general health is either good, very good, or excellent.

Self efficacy scores ranged from 1 to 5 with a mean score of 4.3, a median score of 4.2, and a standard deviation of 0.68. STOFHLA scores ranged from 2 to 36 with a mean score of 32, a median of 34, and a standard deviation of 6.3. Additionally, 88% of the population scored at the adequate health literacy level.

Table 2 shows that education and acculturation most influence STOFHLA scores compared to either the location within Bexar County or ethnicity. People with less than a high school education are most likely to have low levels of health literacy (27%), followed by those who have a high school education only (13%).

Table 3 presents STOFHLA scores according to each Bexar County sector. This table shows that, among this sample, there are a considerable number of people who live in the southern areas of the county with inadequate health literacy. Thirteen percent of the people who live in the West sector and 15% of those who live in the South sector have low levels of health literacy. However, respondents with inadequate health literacy are not limited to these geographic areas but are spread throughout the county.

When examining health literacy scores among different ethnic groups, Mexican Americans (15 percent) and people of “other” ethnic groups (14 percent) have the highest rates of inadequate health literacy (see Table 4). 

Language use has the largest effect on health literacy. People who speak only Spanish or Spanish more than English are more likely to have inadequate health literacy (see Table 5).

Even among those who speak Spanish and English equally, almost 16% have difficulty with health literacy.

Conclusions

The majority of people who participated in this study had adequate health literacy. Contrary to expectations at the beginning of the project, over 85% of the general population has adequate health literacy. However, there still is a significant proportion of the population who scored in the marginal and inadequate range. People with low health literacy scores are geographically spread throughout the city, but they seem to share certain characteristics which are supportive of previous literature including being part of a minority group, having less education, and having fewer financial resources. Additionally, these people were more likely to be Spanish speakers.

Prior to this study, health literacy levels had not been measured in any Texas city or county. This new data shows the extent of the problem in Bexar County. Increased knowledge of the prevalence of low health literacy in Bexar County, through the efforts of SAHLI, should help raise awareness about the importance of this issue and encourage interventions to start addressing the health disparities that are strongly linked to education. If the community has evidence that a significant proportion of the population has low health literacy, it may help motivate systems to strive to be more accessible and to provide a more user-friendly environment for our citizens, e.g., one that is easier to navigate. For example, pictograms or symbols might be used to help patients find their way in healthcare facilities. Health forms should be made easier to read and complete. Instead of being be written at a 12th -grade level, Web sites and educational pamphlets should be written at a 5th- or 6th-grade level to facilitate comprehension. Health care providers should also use more common terminology when talking to patients. Addressing health literacy can become a priority which, we hypothesize, would lead to a healthier community.

It is important to recognize that there are limitations to this study. First, the sample was selected using a geographically stratified convenience strategy. Thus, care must be taken when interpreting these results due to the non-random nature of the sample selection. Second, there may be selection bias as well because people who have literacy problems may have been less likely to agree to participate in this survey. Nevertheless, even with these limitations, the data that were collected provide an important glance into this pressing public health problem.