Test-retest reliability of a complex human immunodeficiency virus research questionnaire administered by an Audio Computer-assisted Self-interviewing system

A social worker intervention to reduce post-hospital mortality in HIV-infected adults in Tanzania (Daraja): Study protocol for a randomized controlled trial

BACKGROUND

In sub-Saharan Africa (SSA), hospitalized HIV-infected patients who are discharged home have been shown to experience extremely high mortality rate. Daraja is an individual-level, time-limited, five-session case management intervention aiming to link hospitalized HIV-infected patients to outpatient HIV care upon discharge.

METHODS

A randomized control trial will aim at evaluating the efficacy of Daraja intervention on reducing mortality in hospitalized HIV-infected patients upon discharge from hospital. The study will recruit 500 hospitalized HIV-infected adults who are ART naïve or defaulted for >7 days from hospitals in Mwanza region, Tanzania. Participants will be enrolled during hospitalization and a baseline assessment will be done. Participants will be randomized to receive either the standard of care HIV linkage, or the Daraja intervention a day before the expected hospital discharge date. The Daraja intervention includes five sessions delivered by a social worker over a 3-month period. All participants will complete follow-up assessment at month 12 and 24. Measures will include 1-year survival, HIV care continuum outcomes (linkage, retention, antiretroviral adherence, and viral suppression), and cost (incremental cost of the intervention and cost per life saved). Quality assurance data will be collected, and the feasibility and acceptability of the intervention will be described. Statistical analysis will assess the effectiveness of the Daraja intervention on improving survival and HIV care continuum outcomes.

DISCUSSION

Hospitalized HIV-infected patients who are being discharged home have higher mortality due to poor linkage to primary HIV care. The Daraja intervention has the potential to address barriers that prevent successful transition from hospital to primary HIV care.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT03858998. Registered on 01 March 2019.

Validity of Measures Assessing Oral Health Beliefs of American Indian Parents

OBJECTIVES

This aimed to validate measures of constructs included in an extended Health Belief Model (EHBM) addressing oral health beliefs among American Indian (AI) parents.

METHODS

Questionnaire data were collected as part of a randomized controlled trial (n = 1016) aimed at reducing childhood caries. Participants were AI parents with a preschool-age child enrolled in the Navajo Nation Head Start program. Questionnaire items addressed five EHBM constructs: perceived susceptibility, severity, barriers, benefits, and parental self-efficacy. Subscales representing each construct underwent reliability and validity testing. Internal consistency reliability of each subscale was evaluated using Cronbach's alpha. Convergent validity was assessed using linear regression to evaluate the association of each EHBM subscale with oral health-related measures.

RESULTS

Internal consistency reliability was high for self-efficacy (α = 0.83) and perceived benefits (α = 0.83) compared to remaining EHBM subscales (α < 0.50). Parents with more education (p < 0.0001) and income (p = 0.0002) perceived dental caries as more severe younger parents (ps = 0.02) and those with more education (ps < 0.0001) perceived greater benefits and fewer barriers to following recommended oral health behavior. Female parents (p < 0.0001) and those with more education (p = 0.02) had higher levels of self-efficacy. Parental knowledge was associated with all EHBM measures (ps < 0.0001) excluding perceived susceptibility (p > 0.05). Parents with increased self-efficacy had greater behavioral adherence (p < 0.0001), whereas lower behavioral adherence was associated with parents who reported higher perceived barriers (p < 0.0001). Better pediatric oral health outcomes were associated with higher levels of self-efficacy (p < 0.0001) and lower levels of perceived severity (p = 0.02) and barriers (p = 0.05).

CONCLUSIONS

Results support the value of questionnaire items addressing the EHBM subscales, which functioned in a manner consistent with the EHBM theoretical framework in AI participants.

Assessment of parental oral health knowledge and behaviors among American Indians of a Northern Plains tribe

OBJECTIVES

Investigate the relationship between sociodemographic variables and oral health knowledge and behaviors of American Indian (AI) parents as the initial step in a program aimed at reducing caries experience among AI children.

METHODS

Survey data were collected from a sample of 147 AI parents of children ages 0-7 years who are residents of a Northern Plains reservation. Questions addressed sociodemographic variables for parents/their children and parent oral health knowledge and behavior. Overall knowledge was measured as percentage of items answered correctly. Overall behavior was measured as percentage of items reflecting behavior consistent with accepted oral health recommendations. Oral health knowledge and behaviors, and the relationship between them, were evaluated across groups defined by quartiles.

RESULTS

Parent sociodemographic variables were not significantly associated with behavior scores. Female gender, higher level of education, and higher income were significantly and positively associated with mean knowledge scores. Behavior and knowledge scores were significantly correlated. On average, survey participants identified the best answer for 75 percent of knowledge items and engaged in 58 percent of optimal oral health behaviors. Participants in higher oral health knowledge quartiles had greater adherence with recommended oral health behaviors than those in lower quartiles.

CONCLUSIONS

Surveyed AI parents had reasonably high levels of knowledge about oral health and caries prevention for their children but engaged at relatively lower levels in parental behaviors necessary to promote oral health. Strategies focused on behavior change, rather than knowledge alone, may be most likely to affect oral health outcomes for AI children.

Test-retest reliability and predictors of unreliable reporting for a sexual behavior questionnaire for U.S. men

Accurate knowledge about human sexual behaviors is important for increasing our understanding of human sexuality; however, there have been few studies assessing the reliability of sexual behavior questionnaires designed for community samples of adult men. A test-retest reliability study was conducted on a questionnaire completed by 334 men who had been recruited in Tucson, Arizona. Reliability coefficients and refusal rates were calculated for 39 non-sexual and sexual behavior questionnaire items. Predictors of unreliable reporting for lifetime number of female sexual partners were also assessed. Refusal rates were generally low, with slightly higher refusal rates for questions related to immigration, income, the frequency of sexual intercourse with women, lifetime number of female sexual partners, and the lifetime number of male anal sex partners. Kappa and intraclass correlation coefficients were substantial or almost perfect for all non-sexual and sexual behavior items. Reliability dropped somewhat, but was still substantial, for items that asked about household income and the men's knowledge of their sexual partners' health, including abnormal Pap tests and prior sexually transmitted diseases (STD). Age and lifetime number of female sexual partners were independent predictors of unreliable reporting while years of education was inversely associated with unreliable reporting. These findings among a community sample of adult men are consistent with other test-retest reliability studies with populations of women and adolescents.

Development, Reliability, and Validity of an Urban Trail Use Survey

Purpose.

To evaluate the psychometric characteristics of the Research on Urban Trail Environments (ROUTES) Trail Use Questionnaire.

Design.

Test-retest reliability was assessed by repeated measures (study 1); validity was assessed by comparing reported trail use to self-reported and objectively measured physical activity (PA) levels (study 2).

Setting.

Study 1: a religious institution situated near a Los Angeles trail. Study 2: 1-mile buffer zones surrounding three urban trails (Chicago, Dallas, and Los Angeles).

Subjects.

Thirty-four adults between 40 and 60 years of age (10 men and 24 women) completed the ROUTES questionnaire twice (study 1). Study 2 participants were 490 adults (48% female and 73% white), mean age 48 years.

Measures.

Trail use for recreation and transportation purposes, time and distance spent on trails, and characteristics of the trail and other trail users. PA was measured using the International Physical Activity Questionnaire and accelerometry.

Analyses.

Pearson correlation coefficients and κ statistics were used for test-retest reliability for continuous and categorical variables, respectively. Generalized linear models were used to evaluate hypotheses on PA comparing trail users and nonusers.

Results.

Test-retest statistics were acceptable (κ = .57, r 5 .66). Validity was supported by correlations between indices of trail use with self-reported PA and accelerometry, and significant group differences between trail users and nonusers in PA levels.

Conclusions.

The ROUTES Trail Use Questionnaire demonstrated good reliability and validity.

Test-retest reliability of a sexual behavior interview for men residing in Brazil, Mexico, and the United States: the HPV in Men (HIM) Study

Understanding the natural history of sexually transmitted infections requires the collection of data on sexual behavior. However, there is concern that self-reported information on sexual behavior may not be valid, especially if study participants are culturally and linguistically distinct. The authors completed a test-retest reliability study of 1,069 men recruited in Brazil, Mexico, and the United States in 2005 and 2006. All of the men completed the same computer-assisted self-interview approximately 3 weeks apart. Refusal rates, kappa coefficients, and intraclass correlation coefficients were calculated for the full sample and by country, age, and lifetime number of female sex partners. Reliability coefficients for each study site and the combined population were high for almost all questions. With few exceptions, the authors found high test-retest reliability with a computer-assisted self-interview on sexual behavior used in 3 culturally and linguistically distinct countries.

Implementation of audio computer-assisted interviewing software in HIV/AIDS research

Computer-assisted interviewing (CAI) has begun to play a more prominent role in HIV/AIDS prevention research. Despite the increased popularity of CAI, particularly audio computer-assisted self-interviewing (ACASI), some research teams are still reluctant to implement ACASI technology because of lack of familiarity with the practical issues related to using these software packages. The purpose of this report is to describe the implementation of one particular ACASI software package, the Questionnaire Development System (QDS; Nova Research Company, Bethesda, MD), in several nursing and HIV/AIDS prevention research settings. The authors present acceptability and satisfaction data from two large-scale public health studies in which they have used QDS with diverse populations. They also address issues related to developing and programming a questionnaire; discuss practical strategies related to planning for and implementing ACASI in the field, including selecting equipment, training staff, and collecting and transferring data; and summarize advantages and disadvantages of computer-assisted research methods.

Psychological and behavioral correlates of entering care for HIV infection: the Antiretroviral Treatment Access Study (ARTAS)

The present study sought to examine psychological and behavioral variables as predictors of attending an HIV medical care provider among person's recently diagnosed with HIV. The study, carried out between 2001 and 2003, was a two-arm randomized intervention trial with participants recruited from public HIV testing centers, sexually transmitted disease (STD) clinics, hospitals, and community-based organizations in Atlanta, Georgia; Baltimore, Maryland; Miami, Florida; and Los Angeles, California. Eighty-six percent of those enrolled (273) had complete baseline and 12-month follow-up data. Measures of number of months since HIV diagnosis, readiness to enter care (based on stages of change), barriers and facilitators to entering care, drug use, and intervention arm (case managed versus simple referral) were examined as predictors of attending an HIV care provider, defined as being in care at least once in each of two consecutive 6-month follow-up periods. In logistic regression, seeing a care provider was significantly more likely among participants diagnosed with HIV within 6 months of enrollment (odds ratio [OR] = 2.52, 95% confidence interval [CI], 1.25, 5.06), those in the preparation versus precontemplation stages at baseline (OR = 2.87, 95% CI, 1.21, 6.81), those who reported at baseline that someone (friend, family member, social worker, other) was helping them get into care (OR = 2.13, 95% CI, 1.02, 4.44), and those who received a case manager intervention (OR = 2.16, 95% CI, 1.23, 3.78). The findings indicate a need to reach HIV-positive person's soon after diagnosis and assist them in getting into medical care. Knowing a person's stages of readiness to enter care and their support networks can help case managers formulate optimal client plans.

The feasibility of audio computer-assisted self-interviewing in international settings

OBJECTIVE

To determine the feasibility of using audio computer-assisted self-interviewing (ACASI) for data collection in developing countries, and to compare responses to questions eliciting sensitive information about sexual behavior using ACASI versus computer-assisted personal interviewing (CAPI) in five developing countries.

DESIGN

A feasibility study determined whether ACASI could be used in populations in developing countries. A follow-up, randomized crossover study compared responses to questions eliciting sensitive information about sexual behavior using ACASI versus CAPI.

METHODS

The NIMH Collaborative HIV/STD Prevention Trial conducted a feasibility study of ACASI in convenience samples in China, India, Peru, and Russia, then a randomized crossover ACASI versus CAPI study among volunteers in these countries plus Zimbabwe.

RESULTS

Approximately equal numbers of men and women completed the feasibility study; the results suggested a high comfort level among participants. Married respondents in China and India appeared to give unreliable responses on sexual activity. In the crossover study, the pattern of responses to sensitive questions showed few differences. In China, higher rates of sexual risk were reported on CAPI. In Peru and Russia, differences by mode were found in the number of partners in the past year.

CONCLUSION

Despite variable computer experience and literacy, feasibility study participants reported ease in completing ACASI, and preferred a computer to an interviewer for answering sensitive questions, or had no preference. In the crossover study, most participants gave similar responses on both modes of survey administration. ACASI appears to be feasible in these settings, although low literacy may pose problems if participants cannot clarify questions.

Factors associated with use of HIV primary care among persons recently diagnosed with HIV: Examination of variables from the behavioural model of health-care utilization

The delay between testing positive for human immunodeficiency virus (HIV) and entering medical care can be better understood by identifying variables associated with use of HIV primary care among persons recently diagnosed with the virus. We report findings from 270 HIV-positive persons enrolled in the Antiretroviral Treatment Access Study (ARTAS). 74% had not seen an HIV care provider before enrollment; 26% had one prior visit only. Based on Andersen's behavioural model of health care utilization, several variables reflecting demographic, healthcare, illness, behavioural, and psychosocial dimensions were assessed and used to predict the likelihood that participants had seen an HIV care provider six months after enrollment. Overall, 69% had seen an HIV care provider by six months. In multivariate analysis, the likelihood of seeing a provider was significantly (p<.05) higher among men, Hispanics (vs. non-Hispanic Blacks), those with higher education, those who did not use injection drugs, those with three or more HIV-related symptoms, those with public health insurance (vs. no insurance), and those who received short-term case management (vs. passive referral). The findings support several conceptual categories of Andersen's behavioural model of health services utilization as applied to the use of HIV medical care among persons recently diagnosed with HIV.

Perceptions of Human Immunodeficiency Virus (HIV) Testing Services Among HIV-Positive Persons Not in Medical Care

BACKGROUND AND OBJECTIVES

Human immunodeficiency virus (HIV) counseling, testing, and referral (CTR) are provided in a wide variety of settings.

GOAL

To compare, by test setting, the perceptions of the testing experience among HIV-positive persons who were not receiving medical care.

DESIGN

A baseline questionnaire was administered at enrollment into the Antiretroviral Treatment Access Study by the use of audio computer-assisted self-interview.

RESULTS

Of 316 respondents, 27% reported that the counselor did not spend enough time with them and 22% that the counselor did not answer all questions. The odds were higher that persons in the following settings, compared with those at HIV test sites, would report that the counselor did not spend enough time with them: office of private physician or health maintenance organization (HMO) (adjusted odds ratio [AOR], 5.24; 95% confidence interval, 1.26-21.7), jail (AOR, 5.10; 95% CI, 1.06-24.6), and emergency room (ER) or hospital overnight visit (AOR, 2.93; 95% CI, 1.15-7.44). Similarly, the odds were higher that persons in the following settings compared with those at HIV test sites would report that the counselor did not answer all questions: office of private physician or HMO (AOR, 9.62; 95% CI, 2.22-41.7), jail (AOR, 7.87; 95% CI, 1.50-41.4), and ER or hospital overnight visit (AOR, 3.32; 95% CI, 1.11-9.90).

CONCLUSION

Further training and quality assurance in HIV CTR may be needed in some test settings.