Literature documented that numerous factors influence adherence to medication and treatment regimes in populations with chronic illnesses. Factors that hindered adherence and non-adherence were numerous and complex. Factors that inhibit adherence included (to name a few): adverse medication effects, medication complexity (e.g. frequency of dosing, number of medications to be taken), need to eat or not eat within a timeframe of taking medications, lack of disease symptomology, lack of social support, beliefs about health/wellness, concurrent alcohol or illicit drug use, knowledge of the treatment regime, negative provider-patient relationships, and cultural beliefs. Factors that promote adherence included: increased knowledge of the illness and medications, positive attitude toward health/wellness promotion, positive provider-patient relationship, external cues or reminders (e.g. alarms, calendars, pillboxes), established routine for taking medications, family/social support, and sensitivity of care providers to cultural beliefs. ₁, ₂, ₃, ₄, ₅

The rationale and theoretical base for this research came from three areas, blame, harm reduction, and the health belief model. Since adherence is viewed as complex, with numerous factors affecting promoting or inhibiting adherence, it seemed reasonable that no one theory/framework or rationale was applicable.

Blame. Poor adherence has been evident in health promotion regimes and in the management of acute and chronic disease. The impact of poor adherence has been significant and issues have centered on the person as the one to blame. Literature on blame suggested that the person may not solely be responsible for the problem. Characteristics of the primary care provider may contribute to the person's ability or willingness to adhere to treatment regimes. ₆, ₇, ₈ This supported the rationale for measuring locus of control with the Multidimensional Health Locus of Control (MHLC) tool. ₉, ₁₀ This tool extracts information about a person's belief about control of health and chance.

Harm reduction. Harm reduction has been described as an approach with persons with HIV that consist of primary, secondary and tertiary prevention strategies ₁₁ applicable in intervention in a variety of settings that include primary care, the community at large, and in drug abuse programs and systems. Strategies focused on prevention that reduces HIV transmission, access to and retention in drug treatment programs, and HIV education and counseling. Harm reduction additionally utilized the concept of therapeutic communities as a unique modality for impacting HIV. ₁₂ Use of this model supported the utilization of the Psychological Adjustment to Illness Self-Report (PAIS-SR) ₁₃ tool in the areas of (1) health care orientation, (2) domestic environment, (3) sexual relationships, (4) social environment, and (5) psychological distress.

The Health Belief Model. The Health Belief Model (HBM) has been an attempt to explain health behaviors that influence prevention and promotion of personal health. ₁₄ Also the HBM assumed that good health is a goal of most persons and that an person's perception of susceptibility, seriousness, benefits, barriers, and self-efficiency impact the likelihood of a person taking an action that is health related. ₁₅, ₁₆, ₁₇, ₁₈, ₁₉, ₂₀ This model supported the use of both measurement instruments, specifically the PAIS-SR domains of health care orientation, social environment and psychological distress. Use of the MHLC is supported in this model by the degree the respondent believes that his/her health is controlled by self, powerful others, or by fate or luck.

The problem of the lack of specific information regarding selected variables created research questions of interest to this research team. Variables of interest to the researchers in persons who are HIV positive were (1) the effect of psychosocial adjustment to illness, (2) locus of control, and (3) person demographic information. Questions of interest were created by the researchers' interest in any trends that might exist between the variables and adherence to medical treatment regimes in persons who are HIV positive.

RQ1: What psychosocial adjustment variables are associated with low HIV blood viral loads?
RQ2: What psychosocial adjustment variables are associated with high HIV blood viral loads?
RQ3: What are the differences in the psychosocial adjustment variables between the groups with low and high HIV blood viral load levels?
RQ4: What is the locus of control of persons with low and high HIV blood viral load levels?

A non-experimental two-group comparison was used. Advantages of this type of comparison were: (1) it is useful for studying problems in social sciences; (2) it permits the analysis of relationships among a large number of variables in a single study; (3) it allows researchers to analyze how the variables, either singly or in combination, affect the pattern of behavior; (4) it provides information concerning the degree of relationship between the variables that are being studied; and (5) comparisons can be made between two groups on selected dependent variables. ₂₁ Limitations of this design included: 1) a positive association between variables and group membership could result in an assumption that this association is the cause of adherence or non-adherence to medical treatment regimes, and 2) it is just as plausible that the variables have no effect on issues of adherence/non-adherence and that something that has not been measured in this study is effecting adherence/non-adherence. ₂₁

Researchers recruited study participants from the Ryan White Title III project Early Intervention Services (EIS) Clinic located at a large university health sciences center in Oklahoma. The population of this clinic, at the time of this investigation, was: (1) predominately male; (2) age range of 18 to 65 years; (3) exposure categories of homosexual exposure, injection drug use, heterosexual contact, and blood transfusion, with a large number of persons not disclosing their exposure category; (4) varied ethnicity, and (5) HIV status corresponding to AIDS surveillance case definition for adolescents and adults.

After obtaining Human Assurance Board approval, patients of this clinic, who were currently actively participating in highly aggressive antiretroviral therapy (HAART), were listed as potential participants. Subjects were then stratified into two groups based on documentation on the most recent viral load numbers available as reported on their medical records. One group consisted of patients with HIV blood viral loads less than or equal to 400 copies per ml. The second group consisted of patients with HIV blood viral loads equal to or greater than 20,000 per ml. A convenience sample of 60 consenting participants from the systematically stratified randomized groups was used.

First, the staff of the EIS clinic generated a list of subjects from the EIS roster that were currently and actively being treated at the clinic and who are participating in HAART medication regimes. Secondly, subjects from this list were stratified into two groups based on their most recent viral load level. Subjects were then systematically selected by using every n^th person until 30 subjects were in each group. These systematically randomly selected subjects from the list were sent a letter in a sealed envelope that provided the purpose and description of the study.

This letter asked them to call the clinic and report their willingness or unwillingness to participate in the study. Subjects who indicated willingness to participate were given the names of the principal investigator and co-investigator to set up a convenient time for assessment of meeting the study's eligibility criteria, enrollment into the study, and completion of the study's instruments. Subjects who did not respond to the first mailed letter were mailed a second letter. In the case of subjects who did not respond to the second mailed letter or indicated unwillingness to participant in the study the next n^th subject on the list was sent the letter describing the study and asked to call the clinic and indicate willingness or unwillingness to participate in the study. This procedure was done to ensure the most representative sample and the best possible response rate to the mailing. No names of individuals were released to the investigators by the EIS clinic until their willingness to participate and consent for their name to be given to the investigators had been obtained. The principal investigator, co-principal investigator, or collaborators asked all participants to sign the consent form, approved by the Human Assurance Board at the university health sciences center, before any instruments were completed. All research team members received training in administering the instruments. Although all team members received training, the primary investigator collected data on 95% of the participants.

Demographic information sheets and study instruments were coded with identical numbers and a group membership identifier for the purpose of matching scores. Matching on the instruments with the demographic information and group membership was used only for purposes of statistical analysis. No other identifying information was present on any of the forms. Participants were assured of the confidentiality of their identity, HIV status, and scores on the instruments. Data was reported only as group data. No matching of the participants to any demographic data or laboratory values was available to anyone but the principle investigators. All data was entered into a computer program by participant number. Once this was done all identifying information was destroyed. All raw data was kept locked in a file cabinet by the principal investigator.

Sampling Measures/Instruments

Instrument: Psychosocial Adjustment to Illness Scale Self-Report (PAIS-SR).

Instrument: Multidimensional Health Locus of Control (MHLC) Scale-Form C.

The primary difficulty in this study was the willingness of persons in the EIS clinic to participate in the research study. These persons faced many challenges in their day-to-day activities and this might have been an additional burden that they were not willing or able to assume. A monetary incentive ($20.00) was given to each participant to help cover the cost of transportation. Researchers were also willing whenever feasible to schedule appointments conveniently for the participant (e.g. in connection with office visits) All of the data was collected in the EIS clinic in a private examination room. Limitations of the study revolved around the design. These limitations were: (1) positive association of the variables and group membership could result in an assumption of cause of adherence or non-adherence, and (2) an assumption that the variables have no effect on adherence or non-adherence. An additional limitation was the assumption that adherence is actually measured by low HIV blood viral levels. Literature supports that low viral loads maybe related to treatment adherence and other measures (e.g. self report, drug urine testing) are no more valid in assessing adherence. ₃,₆,₈

The authors would like to thanks the members of the research team for their help in planning and carrying out the project. They are:

Sheryl Buckner, MS, RN
Instructor
University of Oklahoma
Health Sciences Center

LaRae I. Huycke, MS, RN, APRN, BC
Instructor
University of Oklahoma
Health Sciences Center

Laurie McGinley, MS, RN
Adjunct Clinical Instructor
University of Oklahoma
Health Sciences Center
College of Nursing

Ann Merrill, MA, MS, RN
Assistant Professor
University of Oklahoma
Health Sciences Center
College of Nursing