Validation of a Hebrew health literacy test

doi:10.1016/j.pec.2007.02.005

Patient Education and Counseling

Volume 67, Issues 1–2, July 2007, Pages 235-239

https://doi.org/10.1016/j.pec.2007.02.005 Get rights and content

Abstract

Objective

To validate a questionnaire measuring health literacy in Hebrew, based on the Short-Test of Functional Health Literacy in Adults (S-TOFHLA).

Methods

Interviews were held with 119 Hebrew speaking patients immediately before undergoing colonoscopy in a gastrointestinal outpatient department in Israel. The Hebrew Health Literacy Test (HHLT) was partly translated from the S-TOFHLA and partly re-written in Hebrew to accommodate for the Israeli health system, language, and culture, and then used to assess respondents’ health literacy levels.

Results

Internal consistency (Cronbach's alpha) of the 12 items representing the health literacy scale was 0.98. Mean score of the health literacy scale was 7.84 ± 5.2 with a range of 0–12. Thirty-five respondents (29.4%) could not correctly answer more than two of the items on the scale (could not read or understand the questions), and 69 (58.0%) answered correctly 11–12 of the items. Those with lower levels of education, women, and the older respondents had significantly lower levels of health literacy. Levels of health literacy as measured with the HHLT and subjective levels of literacy were positively correlated (r_p = 0.671).

Conclusion

The HHLT seems to be a valid measure of health literacy and can be used for assessment of health literacy in Hebrew.

Practice implications

The availability of the test should encourage studies regarding the magnitude of the problem and ways to decrease ill effects of low health literacy in Israel.

Introduction

Numerous studies have indicated that health literacy can influence health status. One possible mechanism may involve ability to understand written instructions and thereby use healthcare services appropriately [1], [2], [3], [4], [5].

In Israel, where the health care system operates in Hebrew, there is a high probability that around 30% of the population has limited Hebrew proficiency; these people include recent immigrants and minority groups whose native language is not Hebrew. Large groups of low literate people frequently pertains in immigrant communities in other countries too. For example, Leyva et al. [6] reported that a high percentage of Spanish-speaking Latino parents in the USA were unable to understand routinely dispensed written medication instructions. Moreover, around 4.5% of the population in Israel have 0–4 years of schooling and another 8.6% have 5–8 years of schooling. Before health literacy can be studied in a non-English speaking community a valid test has to be developed to assess its members’ level of reading ability and health literacy in the language of interest. Such a test does not exist for Hebrew. The most common test in English for measuring health literacy levels is the TOFHLA and S-TOFHLA [7], [8]. Previous tests of reading ability, such as the Rapid Estimate of Adult Literacy in Medicine (REALM), do not measure comprehension [9]. S-TOFHLA measures both numeric literacy and reading and understanding prose passages. Means are also available to test levels of general literacy and assign a grade level, such as the Wide Range Achievement Test-Revised (WRAT-R) [10]. The S-TOFHLA has been translated into Spanish, validated and used in Spanish-speaking communities [11].

For further research regarding this issue in Israel, we validated a test that measures health literacy in Hebrew.

Section snippets

Study population

Patients at a gastrointestinal outpatient department in a large medical center (Rambam Medical Center) in Haifa, Israel, were interviewed while awaiting colonoscopy, patients were not medicated. The patient population comprised 119 outpatients who spoke Hebrew.

All patients scheduled for colonoscopy on random days of the week, between March 2004 and February 2005, were asked for consent to be interviewed. The response rate was high (89%): four patients refused to be interviewed and 15 patients

Results

Patients’ mean age was 55.9 years with a standard deviation of 15.1; 30% were older than 65, 48% were men, and 52% were women. Fifty-two percent were born in Israel. Most of the respondents were married (76%) and had children (87%). Eleven percent had less than 12 years of schooling, 33.5% had a high-school education and 55.5% reported having education above high school.

Between 64 and 71% of the respondents answered correctly each of the four items measuring numeracy; 26.9% of the respondents

Discussion

This study is the first to develop, validate, and use a HHLT for Hebrew speakers and readers based on the S-TOFHLA. The Israeli health system operates mostly with Hebrew-speaking healthcare providers. Therefore, if research on effective ways to promote knowledge and health behavior is warranted, an instrument that can measure health literacy in the local language (Hebrew) should be developed. In Hebrew, there is no gold standard that measures literacy comparable to the tests that are available

References (14)

L.D. Chew et al.
The impact of low health literacy on surgical practice
Am J Surgery
(2004)
D.H. Howard et al.
The impact of low health literacy on the medical costs of Medicare managed care enrollees
Am J Med
(2005)
M. Leyva et al.
Health literacy among Spanish-speaking Latino parents with limited English proficiency
Ambulatory Pediatr
(2005)
D.W. Baker et al.
Development of a brief test to measure functional health literacy
Patient Educ Couns
(1999)
A.C. Aguirre et al.
Performance of the English and Spanish S-TOFHLA among publicly insured Medicaid and Medicare patients
Patient Educ Couns
(2005)
L.D. Chew et al.
The impact of low health literacy on surgical practice
Am J Surg
(2004)
M.V. Williams et al.
Inadequate literacy is a barrier to asthma knowledge and self care
Chest
(1998)

There are more references available in the full text version of this article.

Cited by (38)

Does acculturation narrow the health literacy gap between immigrants and non-immigrants—An explorative study
2017, Patient Education and Counseling
Citation Excerpt :
The S-TOFHLA was administered in the respondent’s native language and thus constitutes the language-independent measure (LIM). Even though originally developed in the US [24], the S-TOFHLA has been adapted and validated in a number of other linguistic contexts, such as Israel [37], Turkey [38], as well as Switzerland [39]. The three BHLS asked participants in their respective native languages about their confidence and capability to read, understand and fill out medical information: (a) “How confident are you filling out medical forms by yourself?”
To compare functional health literacy (HL) levels in three immigrant groups to those of the German- and Italian-speaking non-immigrant population in Switzerland. Moreover, to investigate whether language-independent, respectively language-dependent, functional HL and variables of acculturation were associated with self-reported health status among immigrants.
Language-independent HL was assessed with the Short Test of Functional Health Literacy (S-TOFHLA) in the respective native languages. Language-dependent HL was measured using Brief Health Literacy Screeners (BHLS) asking about participants’ confidence in understanding medical information in the language of the host country. Measures of acculturation included length of stay and age when taking residency in Switzerland.
In particular Albanian- and Portuguese-speaking immigrants had lower levels of functional HL. In unadjusted analysis “age when taking residency in Switzerland” was associated with the BHLS. Adjusted analysis showed that the BHLS were significantly associated with self-reported health among all immigrant groups (p ≤ 0.01).
Functional HL that is dependent on understanding of medical information in the language of the new host country is a better predictor for self-reported health status among immigrants than language-independent HL.
In the clinical setting limited functional HL might be a significant obstacle to successful disease treatment and prevention in immigrants.
State of the science of health literacy measures: Validity implications for minority populations
2015, Patient Education and Counseling
To review the evidence supporting the validity of health literacy (HL) measures for ethnic minority populations.
PubMed, CINAHL, and PsycINFO databases were searched for HL measures between 1965 and 2013.
A total of 109HL measures were identified; 37 were non-English HL measures and 72 were English language measures. Of the 72 English language measures, 17 did not specify the racial/ethnic characteristic of their sample. Of the remaining 55 measures, 10 (18%) did not include blacks, 30 (55%) did not include Hispanics, and 35 (64%) did not include Asians in their validation sample. When Hispanic and Asian Americans were included, they accounted for small percentages in the overall sample. Between 2005–2013, a growing number of REALM and TOFHLA translations were identified, and new HL measures for specific cultural/linguistic groups within and outside the United States were developed.
While there are a growing number of new and translated HL measures for minority populations, many existing HL measures have not been properly validated for minority groups.
HL measures that have not been properly validated for a given population should be piloted before wider use. In addition, improving HL instrument development/validation methods are imperative to increase the validity of these measures for minority populations.
Optimal health literacy measurement for the clinical setting: A systematic review
2015, Patient Education and Counseling
To identify the optimal measurement instrument for assessing health literacy in a clinical setting.
Seven databases were searched for studies evaluating health literacy instruments used with patients. Standardised systematic review methods were used by two reviewers independently assessing eligibility, extracting data and evaluating study quality. A narrative summary was produced.
The searches identified 626 articles of which 64 were eligible. Forty-three different health literacy instruments were identified. The quality of these instruments, based on their psychometric properties, varied considerably. The majority of health literacy instruments were found to only assess communicative health literacy of which the numeracy element was often not represented. The NVS instrument was found to be the most practical health literacy instrument to use.
There is an urgent need to develop and psychometrically test a more encompassing health literacy instrument applicable in clinical settings as well as health promotion in general.
In the absence of a more comprehensive health literacy instrument, the NVS is a practical instrument to quickly assess for health literacy in a clinical setting.
Cross-cultural adaptation and validation of the health literacy assessment tool METER in the Portuguese adult population
2014, Patient Education and Counseling
Citation Excerpt :
Health literacy is commonly measured using instruments based on word recognition or pronunciation: Medical Term Recognition Test (METER) [10], Rapid Estimate of Adult Literacy in Medicine (REALM) [11], Short Assessment of Health Literacy for Spanish-speaking Adults (SAHLSA) [12], Medical Terminology Achievement Reading Test (MART) [13]; or reading comprehension and numeracy: Newest Vital Sign (NVS) [8], Test of Functional Health Literacy in Adults (TOFHLA) [14]. Most instruments were initially developed in English or Spanish and are being adapted worldwide [15–19]. Word pronunciation-based instruments perform well in English but have failed adaptation to languages with very high letter to sound correspondence (such as Spanish, Portuguese and Korean) because they are unable to discriminate between health literacy and ability to read [12,16,20].
We aimed to culturally adapt and validate METER in the Portuguese population, and to define cut-off values for adequate health literacy.
We used the standard procedure for the adaptation of the words and surveyed health professionals to select the non-words. The instrument was administered to a total sample of 249 participants and retested in a sub-sample of 45 after three months. Cut-offs were defined using the modified Angoff procedure. Construct validity was assessed through association with educational attainment and health-related occupation.
Exploratory factor analysis revealed two dimensions of the instrument, one for words and another for non-words. METER showed a high degree of internal consistency, and acceptable test–retest reliability. Adequate health literacy was defined as scoring at least 35/40 in words and 18/30 in non-words. Physicians scored higher than any other group, followed by health researchers, researchers from other areas and by people with progressively lower levels of education (p < 0.001).
We culturally adapted a brief and simple instrument for health literacy assessment, and showed it was valid and reliable.
The Portuguese version of METER can be used to assess health literacy in Portuguese adults and to explore associations with health outcomes.
Functional health literacy in Switzerland-Validation of a German, Italian, and French health literacy test
2013, Patient Education and Counseling
Citation Excerpt :
Currently, validated S-TOFHLAs only exist in a limited number of languages. It has been translated and validated into Spanish [6], Brazilian Portuguese [24], Hebrew [25], Serbian [21], Chinese (Cantonese) [27], and Mandarin [28]. Carthery-Goulart et al. [24] translated the S-TOFHLA into Portuguese and adapted it to the Brazilian context to convey information about the Brazilian health system.
This study aimed to translate and validate German, Italian, and French versions of the Short-Test of Functional Health Literacy (S-TOFHLA), to be used in Switzerland and its neighboring countries.
The original English version of the S-TOFHLA was translated by applying standardized translation methods and cultural adaptations. 659 interviews were conducted with Swiss residents in their preferred language (249 German, 273 Italian, and 137 French). To assess the validity of the measures, known predictors for health literacy (age, education, and presence of a chronic condition) were tested.
For all three language versions, results show that younger participants, participants with a higher education and participants with chronic medical conditions had significantly higher levels of health literacy. Furthermore, the three health literacy scales categorized participants into three health literacy levels with most people possessing either inadequate or adequate levels. The highest levels of health literacy were found in the Swiss-German sample (93%), followed by the Swiss-French (83%) and Swiss-Italian (67%) samples.
The German, Italian, and French versions of the S-TOFHLA provide valid measures of functional health literacy.
The translated versions can be used in the three different language regions of Switzerland as well as in neighboring countries following ‘country specific’ adjustments and validations.
Pharmacotherapy literacy level and predictors of low literacy among diabetes mellitus type 2 patients in Serbia
2023, BMC Public Health

View all citing articles on Scopus

View full text

Short communicationValidation of a Hebrew health literacy test

Abstract

Objective

Methods

Results

Conclusion

Practice implications

Introduction

Section snippets

Study population

Results

Discussion

Am J Surgery

Am J Med

Ambulatory Pediatr

Patient Educ Couns

Patient Educ Couns

Am J Surg

Inadequate literacy is a barrier to asthma knowledge and self care

Chest

Short communication
Validation of a Hebrew health literacy test