Performance of the English and Spanish S-TOFHLA among publicly insured Medicaid and Medicare patients

https://doi.org/10.1016/j.pec.2004.03.007Get rights and content

Abstract

The objective of this study was to analyze the psychometric properties of the English and Spanish S-TOFHLA for three patient populations of interest. Despite being widely used, little is written about psychometric properties of the S-TOFHLA, particularly the Spanish version. Participants were Medicaid and Medicare patients: 936 non-Hispanic and 368 Hispanic patients completed the English S-TOFHLA; 1066 Hispanics completed the Spanish S-TOFHLA. Validity of both versions of the S-TOFHLA was supported by strong positive relationships with education and inverse relationships with age. Significant differences between scores for men and women remained after adjusting for level of education. Score differences occurred across numerous items. More research is needed on the S-TOFHLA and other health literacy instruments to determine why women score differently than men. Variability in literacy skills within subgroups of patients highlights the importance of healthcare providers being sensitive to patients’ literacy levels in both spoken and written communications.

Introduction

According to the National Assessment of Adult Literacy (NALS), low literacy is a problem in the United States, particularly among disadvantaged and minority populations [1]. Functional literacy is particularly crucial in the domain of health. More than ever, patients must understand and process a complex array of health information, such as taking multiple prescriptions with different doses, handling self-administered medical equipment such as nebulizers and inhalers, and completing consent forms related to their health care [2]. Moreover, several studies have shown that many materials, such as patient education brochures, consent forms, and discharge papers, are written at a level that exceeds the reading ability of most of the American population [3].

Health literacy is associated with important health outcomes and behaviors, such as decreased preventive care among Medicaid beneficiaries [4], increased risk of hospital admission [2], increased risk of gonorrhea [5], lower rates of cervical cancer screening [6], worse diabetes care [7], [8], a lower incidence of breastfeeding [9], worse functional health [10], and lower patient satisfaction [11]. To operationalize literacy, several literacy instruments have been developed. The test of functional health literacy in adults (TOFHLA) is widely used to test patient comprehension of health-related materials [12]. Alternatives are the rapid estimate of adult language in medicine (REALM), a word recognition and pronunciation exercise [13], and the wide range achievement test-revised (WRAT-R), an arithmetic, spelling, and reading assessment not specific to medicine. The TOFHLA differs from other literacy instruments in that it is available in both English and Spanish, and in several different lengths [14], [15]. Little research exists on different versions of the TOFHLA, and little is known about how the shortened version performs on an item level, and among different populations. Similarly, little is known about performance of the Spanish TOFHLA, which was developed by translating the English TOFHLA into Spanish, and then back translating it into English [16]. In a single sample study correlations between the English and Spanish TOFHLA demonstrate consistency between the two, but performance of the Spanish TOFHLA has not yet been examined to the degree that it has for the English TOFHLA [16]. In particular, we were interested in comparing how the English and Spanish S-TOFHLA performed in samples of Medicaid and Medicare patients.

Accordingly, the objectives of our study were to analyze the English and Spanish S-TOFHLA for three patient populations of interest: non-Hispanics who took the S-TOFHLA in English, Hispanics who took the S-TOFHLA in English and Hispanics who chose to take the S-TOFHLA in Spanish. Within each patient group we examined: (1) overall and item performance; (2) performance for subgroups defined by education level, sex, and age; (3) age, sex, and ethnicity subgroup performance when stratifying by education; and (4) summary item-level data for the aforementioned subgroups. Based on our own and others’ prior work, we hypothesized there would be a significant relationship between literacy and education. Although the two constructs are not perfectly correlated, many researchers using a variety of tools have found strong positive relationships [1], [17]. We also hypothesized, as other research has shown, that there would be an inverse relationship between literacy and age [1], [18], [19]. Although women sometimes have higher scores than men on verbal skills assessments [20], [21], within the literacy literature gender is often not related to literacy [22], [23]. Thus, we did not expect sex differences in literacy scores, especially when controlling for years of education.

Section snippets

Instrument

The TOFHLA assesses a patient’s level of comprehension of health related material. It is available in a full-format (a 22-min test, with 50 reading comprehension items in three passages and 17 numeracy items), an abbreviated format (a 12-min test, with 36 reading comprehension items in two passages and four numeracy items) and a shortened version (the S-TOFHLA, a 7-min test, with 36 reading comprehension items in two passages) [12]. The passages on the S-TOFHLA use a modified Cloze procedure

Results

A total of 2370 patients comprise the sample. The English S-TOFHLA was completed by 1304 respondents: 936 self-identified as non-Hispanic (of these 91% were African American) and 368 as Hispanic (93% Puerto Rican, 4% other, 2% Mexican, 1% Dominican, and 1% Cuban). The Spanish S-TOFHLA was completed by 1066 self-identified Hispanics (85% Puerto Rican, 10% Dominican, and 5% other). Demographic characteristics of the respondents in each of the three groups are shown in Table 1. All comparisons

Discussion

The TOFHLA is one of the most widely used instruments to assess health literacy. The shortened TOFHLA is more practical than the longer versions, but less is known about its performance. In this manuscript we present an in-depth analysis of scores on the English and Spanish S-TOFHLA. We compare the performance of Hispanics and non-Hispanics on the English version, and provide the largest sample to date describing item-level performance of the Spanish version.

Four conclusions stand out from our

Practice implications

On the one hand, the results are very encouraging and support the validity of the S-TOFHLA. Our results were similar to those of the 1992 NALS in that less educated and older respondents had lower literacy scores than more educated and younger [1], [18], [19] counterparts. Also, our Spanish-speaking Hispanic respondents were outperformed by African American respondents in this study, similar to the 1992 NALS. Further, our English-speaking Hispanic respondents outperformed African American and

References (27)

  • S.T. Lindau et al.

    The association of health literacy with cervical cancer prevention knowledge and health behaviors in a multiethnic cohort of women

    Am. J. Obstet. Gynecol.

    (2002)
  • D.W. Baker et al.

    Development of a brief test to measure functional health literacy

    Patient Educ. Couns.

    (1999)
  • N.S. Parikh et al.

    Shame and health literacy: the unspoken connection

    Pt. Educ. Couns.

    (1996)
  • Kirsch I, Jungeblut A, Jenkins L, Kolstad A. Adult literacy in America A first look at the results of the National...
  • D.W. Baker et al.

    Functional health literacy and the risk of hospital admission among Medicare managed care enrollees

    AJPH

    (2002)
  • P.P. Lee

    Why literacy matters. Links between reading ability and health

    Arch. Ophthalmol.

    (1999)
  • T.L. Scott et al.

    Health literacy and preventive health care use among Medicare enrollees in a managed care organization

    Med. Care

    (2002)
  • J.D. Fortenberry et al.

    Relation of health literacy to gonorrhoea related care

    Sex Transm. Infect.

    (2001)
  • D. Schillinger et al.

    Association of health literacy with diabetes outcomes

    J. Am. Med. Assoc.

    (2002)
  • M.V. Williams et al.

    Relationship of functional health literacy to patients’ knowledge of their chronic disease. A study of patients with hypertension and diabetes

    Arch. Int. Med.

    (1998)
  • H. Kaufman et al.

    Effect of literacy on breast-feeding outcomes

    S. Med. J.

    (2001)
  • C. Guerra et al.

    Functional health literacy, comorbidity and health status

    J. Gen. Int. Med.

    (2003)
  • J.A. Shea et al.

    Health literacy and patient satisfaction

    J. Gen. Int. Med.

    (2003)
  • Cited by (0)

    View full text