Abstract

We developed eight heavily illustrated booklets covering patients' rights and responsibilities, antiepileptic medicines, and most psychotropic medicines. The language level was very basic but covered a wide range of information. We distributed free copies of the booklets, together with standardized questionnaires, to consumers with and without intellectual disabilities or other developmental disabilities; 604 questionnaires were returned. The majority of consumers indicated that they read the booklets, learned more about their rights/responsibilities and the medicines described, and found the booklets helpful. Consumers with intellectual disability experienced more difficulty than “average” consumers in understanding the materials, but satisfaction and understanding were reportedly high overall. Female and minority respondents indicated somewhat higher satisfaction with the booklets than did their counterparts.

Only about 50% of patients in developed countries who suffer from chronic diseases follow prescribed treatment regimens (World Health Organization, 2003), and as many as one third to one half or more of medical and psychiatric outpatients in the general population do not follow medication regimens (Amdur, 1979; Sackett & Snow, 1979; Van Putten, 1983). This noncompliance (more correctly referred to as nonadherence) has been strongly linked to exacerbation of symptomatology and subsequent re-hospitalization (Coudreaut-Quinn, Emmons, & McMorrow, 1992).

Over the years, it has been well-documented that patient education leads to better adherence to medication regimens and, therefore, to more therapeutic success. For example, Aubert et al. (2003) found that patients with depression who received a series of educational mailings and telephone-counseling calls focusing on depression education, and the importance of medication compliance were significantly more likely to adhere to their medication regimens. Seltzer, Roncari, and Garfinkel (1980) found that following a course of lectures and fact sheets on their illnesses and medications, psychiatric patients adhered better to treatment and were less fearful of side effects and addiction on follow-up. Kelly and Scott (1990) also found that two relatively brief health education interventions (a series of home visits with the patient's family and structured clinic visits with patients and providers) significantly improved medication adherence and quality of life of psychiatric patients.

The need for patient education is especially apparent for people with intellectual and/or developmental disabilities, particularly given that the rate of psychiatric and behavior disorders is much higher in this population than in the general community (Deb, Thomas, & Bright, 2001a, 2001b; Stark, Menolascino, Albarelli, & Gray, 1988). Some 25% to 40% of adults with intellectual disability living in the community take psychotropic medications (Reiss & Aman, 1998, pp. 1, 2; Rinck, 1998). The rates in developmental centers are even higher (Rinck, 1998).

We are aware of one published study in which written information on psychotropic medication was prepared for this population. Strydom and Hall (2001) prepared eight leaflets on various types of medications for use by adults with mild to moderate intellectual disability who had mental illness and were taking psychiatric medications. They found that 24 participants exposed to the booklets did not manifest significantly improved medication knowledge compared with 26 controls (all of whom received psychotropic medicines). Because of the general paucity of written medication information for people in this population, we developed a set of medication education booklets designed to be read (or be read to) and understood by people with intellectual disability/developmental disability. In this paper, we described (a) the contents of the booklets, (b) the overall reception of the booklets by individuals with and without intellectual/developmental disability, and (c) the relationship between consumer characteristics and reported understanding and satisfaction.

In 1998, the U.S. Administration on Developmental Disabilities funded a Training Initiative Project at Ohio State University's Nisonger Center to prepare and disseminate booklets on psychotropic medicines for use by people with developmental and other types of disabilities. The activity was called “Project MED: Medication EDucation for Consumers.” The Nisonger Center had previously hosted an International Consensus Conference on Psychopharmacology and Developmental Disabilities and published a comprehensive book written for professionals, Psychotropic Medications and Developmental Disabilities: The International Consensus Handbook (Reiss & Aman, 1998).

The developmental process entailed the drafting of each booklet with subsequent refinement by consumer advisory boards and by a professional advisory board. The consumer boards typically were comprised of 12 to 14 consumers with (a) intellectual disability; (b) psychiatric illness but no intellectual/developmental disability; and (c) one or two children or adolescents, often with a psychiatric disability. Children and adolescents were included for two reasons. First, although not regarded legally as “majors,” it is still important that young people assent to the use of the medicines that they are taking; this is likely to lead to more effective treatment. Second, psychotropic medicines are being used more and more these days for treating young patients and, like adults, they have a right to factual information about these agents. The consumer boards were tailored to the topic (e.g., taking one or more of the medicines discussed), and they were paid both to read and to comment on the booklets. Feedback was written (notes written directly on the printed booklet) and in the form of group feedback at pizza dinner sessions. The professional advisory board was composed of psychiatrists, pharmacists, psychologists, lawyers, a dentist, nutritionists, patient advocates, pediatricians, and others with special knowledge of psychoactive medicines, individuals with intellectual/developmental disability, and patients' rights.

Our choice of topics was based on the following considerations. First, all medication consumers have a right to information about their rights (including the right to refuse medication for competent persons) and, to enhance the effectiveness of treatment, they should be aware of their responsibilities as good consumers. Second, we wanted to cover all psychoactive compounds commonly taken by consumers with intellectual disability. Collectively, this included all major groups of psychotropic drugs and, because they act directly on the central nervous system, antiepileptic drugs. Occasionally, agents initially developed for physical symptoms (e.g., antihypertensives) have been used to treat behavioral and/or psychiatric symptoms, and we addressed these as well. Finally, various over-the-counter agents (e.g., St. John's wort, megavitamins) have been employed to treat behavioral or psychiatric problems, and we discussed them for this reason.

Over time, we developed eight booklets, all of which were written at the 4th- to 5th-grade reading level and heavily illustrated to help teach key points. Reading level was determined by feedback provided the Microsoft Word readability statistics option, which determines a Flesch-Kincaid Grade Level. Occasionally (e.g., with antipsychotic, responsibility), we defined a complex term with simpler words. When this was done, we did not include the more complex term in our evaluation of reading grade level. The first booklet, Patients' Rights and Responsibilities, covered issues such as basic medical rights (informed consent, right to refuse treatment, right to confidentiality, right to see one's records, and right to continued care). One major section explained in detail how to complain effectively. The sections on patient responsibilities discussed candor with health-care professionals, disclosure of other health problems, the importance of following directions, reporting problems with medicines, and keeping follow-up appointments. The booklet also contained wallet cards that contained suggested questions for patients to pose for their doctors. We obtained sample illustrations from about six local artists. In the end, we chose an illustrator who used a fairly “clean,” straightforward technique that was readily understood. For reasons of cost, all illustrations were in black and white. The illustrations added interest to the booklets and helped to break up blocks of print. We attended to diversity by systematically rotating across ethnicity, age, and gender in depicting patients in these illustrations.

Six booklets were written about groups of psychoactive medicines as follows: (a) Anticonvulsant Medicines, (b) Antipsychotic Medicines, (c) Antidepressant Medicines, (d) Antimanic Medicines, (e) Antianxiety Medicines, and (f) Stimulant Medicines. With the exception of the Anticonvulsant Medicines booklet, which was slightly longer and contained additional sections, all booklets addressed the following topics: (a) members of the drug class, (b) clinical indications, (c) side effects, (d) common dose ranges, (e) common interactions, and (f) issues related to starting and stopping treatment with the medicine. The eighth and last booklet was called Other Behavior Medicines (Blood Pressure Medicines, Naltrexone, Over the Counter). It contained the same topics as the other booklets but, because it covered several groups of substances, each section had to be kept briefer. These booklets ranged from 17 pages (Patient's Rights) to 32 pages (Anticonvulsant Medicines). We felt that although content had to be kept fairly simple for our audience, the volume of information should be as comprehensive as possible. The eight booklets are referenced in Appendix A; Project MED has its own web site: http://psychmed.osu.edu/projectmed.htm.

The Study: Method

In order to evaluate Project MED, we decided to collect data from recipients of the booklets. In the remainder of this article, we describe the survey conducted and the results obtained. We do not know the numbers of booklets with questionnaires that were distributed. If a respondent received more than one medication, it was possible for him or her to submit more than one questionnaire.

Dissemination and Subjects

Subjects

This study was approved by our institutional review board. The review board members agreed that consent was implicit if the subject completed and returned the survey form. As the Project MED booklets were produced, they were disseminated free through a local pharmacy known to have high numbers of clients with intellectual/developmental disability, county boards of mental retardation/developmental disabilities, physicians' offices, a dual diagnosis clinic, and a local children's hospital. The booklets were free; each contained a questionnaire that asked readers to let us know what they thought of the booklet.

Questionnaires

In the postage-paid questionnaires, respondents were promised a payment of \$5 if the questionnaire was mailed to us; respondents were told that responses would be kept confidential.

Demographics

In the booklets respondents were asked a series of demographic questions to enable us to characterize their age, gender, ethnicity, disability status (e.g., intellectual disability, mental illness, learning disability), living arrangement (living alone, with family, group home, supported living, other), and where the booklet was obtained. Two questions were used to determine whether the respondents received a variety of forms of special education and whether they ever received services from a county board of mental retardation/developmental disabilities (a major service delivery mechanism in Ohio). Questions were posed so that we could characterize participants as having or not having intellectual disability.

The actual questions related to the Patients' Rights and the various medication booklets are presented in Tables 3 and 4, respectively, so they are not duplicated here. Basically, the questions focused on “readability” of the text, areas of information covered, and respondents' impressions of how much they learned.

Table 3

Survey Results for Patients' Rights Booklet (N = 163)

Table 4

Survey Results for Booklets 2 Through 7 (Total N = 441)

Results

Combined Sample

A total of 604 responses were returned. The breakdown by booklet type was as follows (with ns in parentheses): Patients' Rights (163), Anticonvulsant Medicines (32), Antipsychotic Medicines (113), Antidepressant Medicines (115), Antimanic Medicines (52), Anxiety Medicines (61), and Stimulant Medicines (68). With the exception of the Stimulant Medicines booklet, which was produced early in the series, these samples reflect when the booklets were produced (large samples associated with the earliest booklets).

There were 456 respondents who reported having a disability, and 372 of those indicated what the disability was: 265 (71.2%) reported an intellectual disability; 89 (23.9%), a physical disability; and 82 (22.0%), a psychiatric disability. Respondents could report more than one disability, so the percentages total more than 100%.

Disentangling Respondents With and Without Intellectual Disability

As noted, in the survey form, we asked a series of questions about presence of disability, receipt of special education, and provision of services from county boards of mental retardation/developmental disabilities. Using these, we constructed a composite intellectual disability category. We classified the sample as having intellectual disability or not based on whether respondents reported certain special education assignments (developmental handicap or multihandicap classroom), receipt of services from a county board of mental retardation/ developmental disabilities, or reported mental retardation as a disability. Of the 604 respondents, 361 (59.8%) were classified as having intellectual disability/developmental disability. We correlated the composite intellectual disability category with all of the other disability categories, and the results appear in Table 1. As there were strong associations between the composite intellectual disability variable and all of the others, we decided to analyze only for presence/absence of composite intellectual disability (rather than these additional categories) when assessing the impact of demographic variables in relation to the types of response given on appraisal of the booklets.

Table 1

Pearson Correlations Between Composite Intellectual Disability and All Other Disability Categories

Characteristics of Subgroups and Combined Group

Characteristics of the respondents by group (intellectual disability, general population, and combined) are summarized in Table 2. Gender was fairly evenly divided for the group with intellectual disability (55% male), whereas the general population sample was largely female (66% female), p = .0001. The racial/ethnic composition was largely White (85%), although the proportions of Hispanic and Asian subjects differed slightly between the subgroups. When we classified the respondents as being either White or minority, there were no differences between the groups (see Table 2). The ethnic composition of the combined samples was highly consistent with that of this county and the state of Ohio. Respondents with intellectual disability predominantly lived in group homes, supported living, or with their families, in this order of frequency. In contrast, the general population participants predominantly lived independently or with their families (group differences were highly significant), p = .0001. About three quarters of respondents considered themselves handicapped, but this was not necessarily confined to intellectual disability/developmental disability. Finally, group ages were as follows: intellectual disability group—M = 40.9, range = 8 to 74, SD = 14.2; no intellectual disability—M = 41.2, range = 14 to 84, SD = 13.2; combined group—M = 41.0 years, range = 8 to 84, SD = 13.8. The age differences (intellectual disability vs. general population) were not significant; very few children or adolescents responded.

Table 2

Project MED Respondent Characteristics

Ratings of Patients' Rights Booklet

The findings for the Patients' Rights booklet appear in Table 3. In general, the responses were surprisingly positive for the evaluation of this booklet. Over 70% of respondents chose the most positive option in response to questions about (a) how much of the booklet they read, (b) whether the booklet was easy to understand, (c) whether they knew more about their rights, (d) whether they knew more about their responsibilities, (e) what to do if they had a problem with their medicine(s), and (f) whether the booklet was considered helpful to them. Conversely, there were areas where the respondents were not as enthusiastic. If one adopts a threshold of less than 60% endorsement for the most positive response possible, then the following questions were answered less positively: (a) how much the respondent learned, (b) whether the pronunciations provided (for nonphonetic, irregular words) helped, and (c) whether the booklets helped respondents think of questions for their doctors or pharmacists. Usually, the second most commonly chosen response was the intermediate choice (e.g., part, a little, some). All in all, it appears that the respondents felt that they obtained considerable knowledge regarding their rights and responsibilities from this booklet.

Ratings of Booklets 2 to 7

The frequency counts for the remaining booklets appear in Table 4. Over 70% of respondents endorsed the most positive choice for the following questions: (a) how much of the booklet that was read, (b) ease of understanding, (c) amount learned about the medicines, (d) amount learned about side effects, (e) what to do if the respondent encountered a problem with the medicine, and (f) helpfulness of the booklet. Once again, there were some areas where respondents were not as positive. If 60% or less on the most positive response option is used to designate areas where respondents were less enthused, the following questions stood out: (a) amount learned from the booklets and (b) whether the booklet helped respondents think of questions for their doctors or pharmacists (see Table 4). Once again, the intermediate choices were the ones that the respondents endorsed second most often. Overall, the survey responses were quite positive.

Patients' Rights: Relation Between Demographics and Positive Responding

Another issue of importance concerns respondent characteristics and differences in survey responses. To assess this, we examined consumer characteristics in relation to the responses given in Tables 3 and 4. The characteristics assessed included gender, age (16–29, 30–39, 40–49, and 50–84 years), ethnicity (African American, Asian American, Hispanic–Non-White vs. White), composite intellectual disability (intellectual disability vs. other), and source of the booklets (drugstore, clinic, county board of mental retardation/developmental disabilities, staff member, and other). Chi-squares and phi coefficients were used to analyze these data, and only findings with a two-tailed probability level of .01 or less are reported for the main (combined group) analyses. Tables summarizing the findings in greater detail can be found on our web site. Because the tables consume several pages, we reported the chi-square and phi coefficients here and the general direction of the association.

Neither age nor gender was correlated with survey responses at the specified alpha level of .01 or less. Race/ethnicity did correlate with taking (or expectation of taking) the medications discussed in the booklets (although White consumers were more likely to be medicated than minority consumers), χ2(1, N = 163) = 7.31, p = .007, ϕ = .21. Presence of intellectual disability (n = 111) versus no intellectual disability (n = 52) was associated with four response variables in a highly predictable manner. The composite intellectual disability versus other comparisons were significant for the following survey questions (see Table 3 for questions): (a) Did you read the booklet by yourself? χ2(1, N = 161) = 70.10, p ≤ .0001, ϕ = .66; (b) How much of the booklet did you read? χ2(1, N = 156) = 13.46, p ≤ .0001, ϕ = .21; (c) How much did you learn from the booklet? χ2(1, N = 162) = 26.39, p ≤ .0001, ϕ = .40; (d) Did the pronunciations make the booklet easier to understand? χ2(2, N = 152) = 9.11, p = .003, ϕ = .24; and (e) Did you learn what to do if you have a problem with your medicine? χ2(1, N = 161) = 15.32, p ≤ .0001, ϕ = .31. In each case, the presence of intellectual disability was associated with less benefit or greater difficulty using the booklets. Nevertheless, the data still indicate that consumers with intellectual disability continued to reap considerable benefit from the booklets.

Finally, the sources of the booklets (drug store, clinic, county board, staff members, and others) were associated with response to nine survey questions (see Table 3). The statistically significant associations were as follows: (a) Did you read the booklet by yourself? χ2(4, N = 159) = 53.94, p ≤ .0001, ϕ = .58; (b) How much of the booklet did you read? χ2(4, N = 154) = 13.84, p = .008, ϕ = .30; (c) Was the booklet easy to understand? χ2(4, N = 160) = 20.86, p ≤ .0001, ϕ = .36; and (d) Did you have help answering the questions? χ2(4, N = 158) = 47.96, p ≤ .0001, ϕ = .55. In each case, consumers who obtained their booklets from a drug store or from other source (as opposed to clinic, county board, or staff member) found the booklets easier to read and to understand. Clearly, the source of booklet question proved to be a proxy for intellectual disability/developmental disabilities. Nevertheless, consumers in the remaining categories continued to show a reasonable command of material contained in the booklet.

Booklets 2 Through 7: Relation Between Demographics and Positive Responses

These analyses were conducted in the same manner as in the previous section, except that the data were taken from the medication booklets (2 through 7). Once again, the findings are keyed to the questions from the survey, in this case from Table 4. There were several significant interactions for gender and expressed satisfaction with the booklets, but examination of the proportion of females in the intellectual disability and no disability groups suggested that there might be an important confound (females made up 44.7% of the intellectual disability group and 65.7% if the no intellectual disability group). Therefore, we also analyzed the data within the intellectual disability group to ensure that any instances of significance could not be ascribed to the fact that there were more females within the general population group. For each analysis, we present the findings for the combined group followed by those for the intellectual disability group only. The findings for gender were as follows: (a) Q1, Did you read the booklet by yourself? χ2(1, N = 438) = 21.46, p ≤ .0001, ϕ = .22; χ2(1, N = 248) = 3.11, p ≤ .08; (b) Q3, Was the booklet easy to understand? χ2(1, N = 437) = 14.04, p ≤ .0001, ϕ = .18; χ2(1, N = 247) = 3.91, p ≤ .05; (c) Q4, How much did you learn from the booklet? χ2(1, N = 434) = 8.52, p = .004, ϕ = .14; χ2(1, N = 247) = 8.11, p = .004; (d) Q7, Did you learn about the side effects of these medicines? χ2(1, N = 436) = 6.64, p = .01, ϕ = .12; χ2(1, N = 247) = 4.47, p = .03; (e) Q12, Was the booklet helpful to you?, χ2(1, N = 437) = 7.10, p = .008, ϕ = .13; χ2(1, N = 247) = 2.37, p = .12; and (f) Q14, Did someone help you answer these questions? χ2(1, N = 438) = 19.88, p ≤ .0001, ϕ = .21; χ2(1, N = 248) = χ2(1) = 3.64, p ≤ .06. Fisher exact probability tests were also computed for 2 × 2 comparisons. In the case of comparisons a and f (Qs 1 and 14), 2-tailed Fisher exact probability tests indicated statistical significance, p ≤ .05, for the intellectual disability group comparisons. In every case, females read more, understood more, and needed less help than did male respondents. However, when the analysis was confined to individuals with intellectual disability, the gender effect became nonsignificant in some instances and generally weaker.

Both composite intellectual disability and source of the booklets were associated with all questions on the survey form. For brevity, we did not summarize each of them here, although they are available in detail on our web site. The associations for intellectual disability (vs. other) were modest to large with the survey responses, ϕ ranged from .16 to .68. As would be expected, consumers with intellectual disability found the booklets more difficult to read and understand as compared with other consumers. The largest effect sizes were found for Q1 (Did you read this booklet by yourself?), ϕ = .68 and Q13 (Did someone help you answer the questions?), ϕ = .63. The relationships between source of booklets and consumer responses were similar to those for intellectual disability/other and responses, with phi coefficients ranging from .18 to .57. The strongest associations were found for Q1 (Did you read the booklet yourself?), ϕ = .57, and Q13 (Did someone help you answer?), ϕ = .47. In each instance, consumers who obtained booklets from drug stores, other, and (sometimes) staff members demonstrated better understanding and knowledge of the booklets. Once again, source of the booklets appears to be a proxy for presence of intellectual disability.

Discussion

About 60% of our respondents could be identified (based on services received) as having intellectual disability. When starting out on this project, we needed to make a basic decision, namely, whether to attempt to provide relatively thorough coverage of patients' rights/responsibilities and the various medicines or to opt for more superficial coverage in the belief that such material would be easier to comprehend. As noted earlier, the process was guided in part by several consumer advisory boards. We believe that we managed to achieve reasonably good coverage of content while also communicating fairly effectively with these consumers. Approximately 88% of consumers reported that they read part or all of Booklet 1 (Patients' Rights), whereas about 91% reported that they read all or part of the medication booklets. About 72% reported that Booklet 1 was easy to understand as compared with 76% for the medication booklets. About 72% of participants responded that both types of booklets were helpful (yes responses marked). All in all, it seems as though those responding at least felt that they obtained some added level of knowledge on the topics covered.

Many authorities (e.g., Schouten & Duckworth, 1999) believe that providing true, informed consent or (in the case of those not competent to consent) assent helps to involve patients and make patients more committed to their treatment. Furthermore, if they have access to information on side effects, drug/drug or drug/food interactions, and commonly used dose ranges, patients will be in a better position to work with their doctors to optimize their treatment. We hope that Project MED has helped to achieve this in some measure both for patients with intellectual disability/developmental disability and for those in the general population.

Not surprisingly, respondents known to be receiving services for intellectual disability encountered more difficulty with some booklet content than did respondents with no intellectual disability. The strength of relationship, as measured by the phi coefficient (ϕ), ranged from quite low (lowest significant ϕ = .16) to moderately high (largest ϕ = .68) (the chi squares and phi coefficients are presented in detail on our web site). The strongest associations for all booklets were found for the questions on whether the respondents read the booklets by themselves and whether they had assistance in answering the questions, all ϕs = .63 to .68. The remaining associations were either very small or modest in size. The smallest significant ϕs for Booklet 1 related to whether the booklet was helpful to the consumer, ϕ = .19, and whether assistance provided for pronouncing nonphonetic words was helpful, ϕ = .21. The smallest significant associations for the remaining booklets occurred for a question as to whether the consumers learned “what to do if you have a problem” with the medicine, ϕ = .16, and whether the booklets prompted the consumers to pose questions of their doctors, ϕ = .23.

Whereas it may have been beneficial to have equal understanding across consumers of all types, this probably would be an unrealistic goal. In a very real sense, this seems to offer face validity to the survey responses, and it suggests that we can have some confidence in the respondents' candor and the accuracy of the responses provided. Furthermore, even though the respondents reported moderate difficulty in actually reading the booklets (many had the booklets read to them) and/or required help filling in the questionnaires, most other group differences were relatively trivial as assessed by the phi coefficient.

For the medication booklets, females were more likely than males to read more, understand more, and require less help for the drug information; but this was not the case for Patient Rights booklets. Although the results were significant for the combined group (Table 6 of summary on our web site, the effect sizes were fairly small, ϕs = .12 to .22. A correlation between gender and composite intellectual disability, r = .21, p < .01, was found, so it is possible that a lower rate of intellectual disability, not a straightforward gender effect, was partially responsible for the differential ratings of the booklets. For the combined groups, 6 of the gender comparisons proved to be statistically significant. when the gender comparisons were confined to the intellectual disability group, however, 3 of the comparisons were significant when tested by chi-square. Nevertheless, as determined by Fisher exact tests, 5 of 6 gender comparisons were confirmed, suggesting consistent trends regardless of group. Our results for gender were in keeping with the bulk of current literature on patient education, in which researchers have reported relatively few differences. For example, Koo, Krass, and Aslani (2005) found that the only difference between genders in evaluating written medicine information to be confined to the design of a package insert (female participants found the inserts to appear less biased than did males). The North Carolina Medication Information Literacy Project (Kirk et al., 2000) found no differences in comprehension by gender alone. However, a Gender × Age interaction did appear: Females (but not males) younger than 50 years of age had higher comprehension scores than did males and females older than 50 years.

In accordance with some of our significant gender findings, Suhonen, Nenonen, Laukka, and Välimäki (2005) in a study in Finland found female gender to be associated with higher importance attached to any information received in a medical setting. In diabetes research, meta-analyses by Brown (1990, 1992) and Padgett, Mumford, Hynes, and Carter (1988) did not show a correlation between effect size of the diabetes educational intervention and the gender of the patient. Thus, although most researchers have not found a definite effect of gender in using patient education materials, those who did usually indicated more favorable evaluation by females. In this respect, the results of gender in Project MED are similar to what is found in the available literature. Unfortunately, we were unable to find any literature of this type for individuals with intellectual disability/developmental disability, indicating a large area of consumer research needing attention in this field.

We had similar difficulty finding any literature focusing on race/ethnicity and learning differences with patient education materials (indeed, we found no reports on gender or ethnicity within the intellectual disability/developmental disabilities literature, suggesting that this area of research has been neglected). Our results showed that minority respondents were more likely than White respondents to learn more about various aspects of the medications and to find the booklets helpful. Although significant (see Table 6 of summary on our web site), the associations were fairly modest in size, ϕs = .13 to .19. Much of the literature touching on this topic in other disciplines, such as cancer (Ramirez & Mondragon, 2002), asthma (Ford, Havstad, Tilley, & Bolton, 1997), computer-based education (Lewis, 1999), and preoperative teaching (Guruge & Sidani, 2002), found no significant differences related to ethnicity. Of the few studies that did find differences, there did not appear to be a consistent pattern in either direction (White vs. minority) in terms of greater knowledge gain. For example, in a study of a breast cancer clinical trial informational video, Curbow, Fogarty, McDonnell, Chill, and Scott (2004) found greater improvement in knowledge scores in non-White women than in White women, even after accounting for lower pretest knowledge. Conversely, Healton, Taylor, Messeri, Weinberg, and Bamji (1999) found that White individuals showed greater gains in knowledge than did minorities after reading a brochure on a therapy to reduce perinatal HIV transmission designed for women with or at risk for HIV infection.

Past inconsistencies in the literature on ethnicity effect appeared to be partially due to the lack of adequate control for education and SES, which are often associated with ethnicity within studies that report ethnicity findings (e.g., Healton et al., 1996; Juarez, Ferrell, & Borneman, 1999; Post, Cegala, & Marinelli, 2001). Unfortunately, the reviewed findings do not elucidate the ethnicity findings within Project MED because the clinical populations are so different. It is possible that the minority subjects in our analyses were struggling with English as a second language and found the simplified language, illustrations, and help with pronunciations to be more useful than did the White respondents. However, Ohio does not have a large non-English-speaking minority population, so that seems unlikely. Nevertheless, replication of this finding would be of considerable practical significance, and future projects of this sort should include some analysis of any differences between ethnic minority and majority participants.

Limitations

There are some obvious limitations to the comparisons that we report here. First, the representativeness of our sample is totally unknown. The respondents could have been characteristic of their respective groups or, at the other extreme, they could have been among the most conscientious and knowledgeable of consumers. Second, some individuals could have been overrepresented because there was nothing to prevent individuals from completing survey forms on more than one booklet type. However, we did have rigorous procedures in place to prevent the same individual from completing more than one survey per booklet. Third, we had to infer identity of the respondent as having intellectual disability or not based on the personal services reported. Fourth, although we summarized the consumers' reported understanding of the information in the booklets, comprehension was not formally measured. Thus, it is possible that there was some error for these figures. Nevertheless, the data provided above suggest that these classifications were valid. Fifth, we did not determine whether readers actually learned anything from the booklets; to do so, we would have had to assess knowledge before and after exposure to them. Finally, we cannot be certain whether the respondents were candid or whether they offered answers that they thought were socially appropriate. However, the instructions were emphatic that the respondents would receive reimbursement even if their reply was that they did not (or would not) read the booklet. In other words, we would have accepted disinclination to read the booklets as a legitimate response and as useful feedback.

Practical Implications

In conducting our literature search, we were struck by the paucity of scientific work on patient educational materials. This appears to be true not only for individuals with intellectual disability, but for most, if not all, medical and psychiatric conditions. As professionals, we may have underestimated our consumers with intellectual disability/developmental disability in the past. In the current survey, most participants both with and without intellectual disability reported that they understood a substantial amount of information on their medications. With this in mind, we encourage consumer advocates to develop similar products on these other matters that are important for succeeding in modern-day life (e.g., managing personal finances, other aspects of health care). In this modern electronic era, it may be appropriate for researchers in future efforts to incorporate a much broader range of media. Also, some agencies expressed interest in the booklets for training direct care staff. Furthermore, the distribution of Project MED materials has been much broader than the developmental disabilities field alone. A number of advocacy organizations concerned with mental illness in the general population have distributed substantial numbers of the series. For consumers, caregivers, and administrators who are interested in using Project MED materials, they can be accessed at our web site http://psychmed.osu.edu/projectmed.htm web page or by contacting either of the two senior authors. With modern technology, there is a wide array of media available for educating consumers: audio, video, computer, iPod, and other modalities. By harnessing these newer—and perhaps more readily understood modes—we may be able to develop educational materials that are broadly understood and perhaps even entertaining.

Table 3

Continued

Table 4

Continued

References

References
Amdur
,
M. A.
1979
.
Medication compliance in outpatient psychiatry.
Comprehensive Psychiatry
20
:
339
346
.
Aubert
,
R. E.
,
G.
Fulop
,
F.
Xia
,
M.
Thiel
,
D.
Maldonato
, and
C.
Woo
.
2003
.
Evaluation of a depression health management program to improve outcomes in first or recurrent episode depression.
American Journal of Managed Care
9
:
374
380
.
Brown
,
S. A.
1990
.
Studies of educational interventions and outcomes in diabetic adults: A meta-analysis revisited.
Patient Education and Counseling
16
:
189
215
.
Brown
,
S. A.
1992
.
Meta-analysis of diabetes patient education research: Variations in intervention effects across studies.
Research in Nursing and Health
15
:
409
419
.
Coudreaut-Quinn
,
E. A.
,
M. A.
Emmons
, and
M. J.
McMorrow
.
1992
.
Adherence and accuracy: Self-medication during inpatient psychiatric treatment.
Journal of Psychosocial Nursing
30
:
32
36
.
Curbow
,
B.
,
L. A.
Fogarty
,
K.
McDonnell
,
J.
Chill
, and
L. B.
Scott
.
2004
.
Can a brief video intervention improve breast cancer clinical trial knowledge and beliefs?
Social Science & Medicine
58
:
193
205
.
Deb
,
S.
,
M.
Thomas
, and
C.
Bright
.
2001a
.
Mental disorder in adults with intellectual disability. 1: Prevalence of functional psychiatric illness among a community-based population aged between 16 and 64 years.
Journal of Intellectual Disability Research
45
:
495
505
.
Deb
,
S.
,
M.
Thomas
, and
C.
Bright
.
2001b
.
Mental disorder in adults with intellectual disability. 2: The rate of behaviour disorders among a community-based population aged between 16 and 64 years.
Journal of Intellectual Disability Research
45
:
506
514
.
Ford
,
M. E.
,
S. L.
,
B. C.
Tilley
, and
M. B.
Bolton
.
1997
.
Health outcomes among African American and Caucasian adults following a randomized trial of an asthma education program.
Ethnicity & Health
2
:
329
339
.
Guruge
,
S.
and
S.
Sidani
.
2002
.
Effects of demographic characteristics on preoperative teaching outcomes: A meta-analysis.
34
:
25
33
.
Healton
,
C.
,
S.
Taylor
,
C.
Burr
,
A.
Dumois
,
N.
Loewenstein
, and
J.
Kaye
.
1996
.
The impact of patient education about the effect of Zidovudine on HIV perinatal transmission: Knowledge gain, attitudes, and behavioral intent among women with and at risk of HIV.
American Journal of Preventative Medicine
12
:(
Suppl 1
).
47
52
.
Healton
,
C.
,
S.
Taylor
,
P.
Messeri
,
G.
Weinberg
, and
M.
Bamji
.
1999
.
Effects of ZDV-based patient education on intentions toward ZDV use, HIV testing and reproduction among a US cohort of women.
AIDS Care
11
:
675
686
.
Juarez
,
G.
,
B.
Ferrell
, and
T.
Borneman
.
1999
.
Cultural considerations in education for cancer pain management.
Journal of Cancer Education
14
:
168
173
.
Kelly
,
G. R.
and
J. E.
Scott
.
1990
.
Medication compliance and health education among outpatients with chronic mental disorders.
Medical Care
28
:
1181
1197
.
Kirk
,
J. K.
,
S.
Krick
,
D.
Futrell
,
T.
Devora
,
S. M.
Caiola
,
E.
Mason
,
W. T.
Sawyer
, and
P.
Gal
.
2000
.
Connecting pharmacy and literacy: The North Carolina Medication Information Literacy Project.
American Journal of Pharmaceutical Education
64
:
277
283
.
Koo
,
M. M.
,
I.
Krass
, and
P.
Aslani
.
2005
.
Patient characteristics influencing evaluation of written medicine information: Lessons for patient education.
Annals of Pharmacotherapy
39
:
1434
1440
.
Lewis
,
D.
1999
.
Computer-based approaches to patient education: A review of the literature.
Journal of the American Medical Informatics Association
6
:
272
282
.
,
D.
,
E.
Mumford
,
M.
Hynes
, and
R.
Carter
.
1988
.
Meta-analysis of the effects of educational and psychosocial interventions on management of diabetes mellitus.
Journal of Clinical Epidemiology
41
:
1007
1030
.
Post
,
D. M.
,
D. J.
Cegala
, and
T. M.
Marinelli
.
2001
.
Teaching patients to communicate with physicians: The impact of race.
Journal of the National Medical Association
93
:
6
12
.
Ramirez
,
T. R.
and
D.
Mondragon
.
2002
.
Computer–based education for patients with cancer at Latino border hospital.
Cancer Nursing
25
:
245
250
.
Reiss
,
S.
and
M. G.
Aman
.
1998
.
Preface.
Psychotropic medications and developmental disabilities: The international consensus handbook (pp. 1–2). Columbus: Ohio State University, Nisonger Center
.
Rinck
,
C.
1998
.
Epidemiology and psychoactive medication.
In S. Reiss & M. G. Aman (Eds.), Psychotropic medications and developmental disabilities: The international consensus handbook (pp. 31–44). Columbus: Ohio State University, Nisonger Center
.
Sackett
,
D. L.
and
J. C.
Snow
.
1979
.
The magnitude of compliance and noncompliance.
In R. B. Haynes, D. W. Taylor, & D. L. Sackett (Eds.), Compliance in health care (pp. 11–22). Baltimore: Johns Hopkins University Press
.
Schouten
,
R.
and
K. S.
Duckworth
.
1999
.
Medicolegal and ethical issues in the pharmacologic treatment of children.
In J. S. Werry & M. G. Aman (Eds.), Practitioner's guide to psychoactive drugs for children and adolescents (2nd ed., pp. 165–181). New York: Plenum
.
Seltzer
,
A.
,
I.
Roncari
, and
P.
Garfinkel
.
1980
.
Effect of patient education on medication compliance.
25
:
638
645
.
Stark
,
J. A.
,
J.
Menolascino
,
M. H.
Albarelli
, and
V. C.
Gray
.
1988
.
Executive summary.
In A. Stark, F. J. Menolascino, M. H. Albarelli, & V. C. Gray (Eds.), Mental retardation and mental health classification, diagnosis treatment services (pp. xi–xviii). New York: Springer-Verlag
.
Strydom
,
A.
and
I.
Hall
.
2001
.
Randomized trial of psychotropic medication information leaflets for people with intellectual disability.
Journal of Intellectual Disability Research
45
:
146
151
.
Suhonen
,
R.
,
H.
Nenonen
,
A.
Laukka
, and
M.
Välimäki
.
2005
.
Patients' informational needs and information received do not correspond in hospital.
Issues in Clinical Nursing
14
:
1167
1176
.
Van Putten
,
T.
1983
.
The clinical management of noncompliance.
In I. Barofsky & R. D. Budson (Eds.), The chronic psychiatric patient in the community: Principles of treatment (pp. 383–395). New York: Spectrum
.
World Health Organization.
2003
.
The magnitude of the problem of poor adherence.
In E. Sabate (Ed.), Adherence to long-term therapies: Evidence for action (pp. 7–9). Geneva, Switzerland: Author. Retrieved June 27, 2006 from http://www.who.int/chronic_conditions/adherencereport/en/
.

Appendix A

References for Project MED Booklets

Aman
,
M. G.
,
B. A.
Benson
,
K. M.
Campbell
, and
B. A.
Haas
.
1999
.
Patients' rights and responsibilities.
Columbus: The Ohio State University. (ISBN 0-9658966-1-7)
.
Aman
,
M. G.
,
B. A.
Benson
,
K. M.
Campbell
, and
B. A.
Haas
.
2000
.
Anticonvulsant medicines (Medicines for people with epilepsy).
Columbus: The Ohio State University. (0-9658966-7-6)
.
Aman
,
M. G.
,
B. A.
Benson
,
K. M.
Campbell
, and
B. A.
Haas
.
1999
.
Antipsychotic medicines.
Columbus: The Ohio State University. (ISBN 0-9658966-2-5)
.
Aman
,
M. G.
,
B. A.
Benson
,
K. M.
Campbell
, and
B. A.
Haas
.
1999
.
Antidepressant medicines.
Columbus: The Ohio State University. (0-9658966-3-3)
.
Aman
,
M. G.
,
B. A.
Benson
,
K. M.
Campbell
, and
B. A.
Haas
.
1999
.
Antimanic medicines (Medicines for people with mood problems).
Columbus: The Ohio State University. (0-9658966-5-X)
.
Aman
,
M. G.
,
B. A.
Benson
,
K. M.
Campbell
, and
H.
Griswold
.
1999
.
Antianxiety medicines.
Columbus: The Ohio State University. (0-9658966-6-8)
.
Aman
,
M. G.
,
B. A.
Benson
,
K. M.
Campbell
, and
H.
Griswold
.
1999
.
Stimulant medicines.
Columbus: The Ohio State University. (0-9658966-4-1)
.
Aman
,
M. G.
,
B. A.
Benson
,
K. M.
Campbell
,
J.
Masty
, and
A.
Rohr
.
2001
.
Other behavior medicines (Blood $bu792$pressure medicines, naltrexone, and over the counter).
Columbus: The Ohio State University. (0-9658966-8-4)
.

A detailed discussion of Project Med can be found at http://psychmed.osu.edu/projectmed.htm

Author notes

Authors:

Michael G. Aman, PhD (aman.1@osu.edu), Professor of Psychology and Psychiatry; Betsey A. Benson, PhD, Adjunct Associate Professor of Psychology; A. Farmer, BA, Graduate Student; and Kristy L. Hall, BA, Graduate Student, Ohio State University, 1581 Dodd Dr., Columbus, OH 43210–1296. Krista M. Malone, MSEd, Children's Bereavement Facilitator, Hospice of Central Ohio, 2269 Cherry Valley Rd., Newark, OH 43055