Kristin A. Riekert

The Beliefs About Medication Scale: Development, Reliability, and Validity

The purpose of the study was to develop and evaluate a psychometrically sound health belief measure, the Beliefs About Medication Scale (BAMS) that can be used with adolescent chronic illness populations whose prescribed treatment includes oral medication. One hundred and thirty-three adolescents (age 11–18 years) with asthma (n = 60), HIV (n = 31), or inflammatory bowel disease (n = 42) completed the BAMS and, along with their parent, a self-report medication interview. A confirmatory factor analysis supported the hypothesized subscales of Perceived Threat, Positive Outcome Expectancy, Negative Outcome Expectancy, and Intent to Adhere to treatment. The subscales evidenced good internal consistency and 3-week test-retest reliability. Univariate and multivariate analyses demonstrated that the health belief constructs accounted for 22% of the variance in medication adherence beyond demographic and illness characteristics. The study provides preliminary evidence of the reliability and validity of a theoretically based measure of health beliefs for adolescents. The BAMS may be a useful tool to evaluate the psychological barriers to adherence that place teenagers at risk for nonadherence.

Understanding and Managing Sampling Issues in Research with Children

Pediatric and clinical child psychologists often conduct research with heterogeneous clinical populations (e.g., children and adolescents with behavior disorders or chronic health conditions) that raise difficult sampling problems. Children and adolescents may be recruited from highly specialized populations and settings, for example, hospitals, that can result in biased samples (Berkson, 1946). Recruiting clinical populations can result in sample self-selection (Betan, Roberts, & McClusky-Fawcett, 1995) and attrition (Aylward, Hatcher, Stripp, Gustafson, & Leavitt, 1985), which may seriously limit generalizability of findings. Finally, in order to reduce heterogeneity of sample characteristics and carefully target their samples, investigators often restrict eligibility criteria, which also limits generalizability.

Ethical Issues in Conducting Research with Pediatric and Clinical Child Populations in Applied Settings

Over and beyond the general ethical principles with which all psychologists should be thoroughly familiar (American Psychological Assoication, 1992; Canter, Bennett, Jones, & Nagy, 1994), research with pediatric and clinical child populations raises special ethical problems that challenge investigators to develop special strategies in order to conduct their research in a responsible manner. The work described in this chapter, which is based on research experiences in hospital, school, and clinic settings, describes a number of knotty ethical problems that have not been well articulated in previous work. These issues include considerations in working with groups that are charged with the oversight of research ethics in clinical settings, problems related to confidentiality of data, obtaining appropriate consent, managing risk-related to psychological vulnerability, for example, depression in adolescents, and maintaining appropriate role boundaries as researchers in clinical settings. Using illustrations from research with varied populations of children and families, we describe a number of ethical issues that have arisen in data collection in clinical settings and consider strategies to manage them. Our aim is to help researchers to anticipate and prevent ethical problems wherever possible and to provide suggestions concerning how their impact on families can be minimized.

Conducting Research with Children and Adolescents in Clinical and Applied Settings

In order to successfully conduct research in clinical and other applied settings with children and adolescents, investigators need to learn to manage a number of logistic problems that can be difficult to anticipate (Drotar, 1989). These problems include developing collaborations with agency and hospital staff that are necessary to recruit subjects (Drotar, 1993), recruiting and maintaining research participants in studies, and managing problems in data collection, especially those that threaten the integrity of study design. Researchers who work with children and families need to anticipate as many of these problems as possible so that they can either implement strategies to prevent them, which is the preferable approach, or develop data analytic approaches to limit their influence on the quality of their data (see Chapter 4, this volume).