Tap 20 — TAPs <<<Documents<<<Home

Computers Link Adolescent Health Research to Rural Settings

Kathryn R. Puskar, Dr. P.H., C.S., R.N., F.A.A.N.
Kathleen Tusaie-Mumford, M.S.N., R.N., C.S.
Jeffrey Rohay, M.S., M.S.I.S.
Jacqueline Lamb, Ph.D., R.N.
Ginger Boneysteele, M.S.N., R.N., C.
Susan Sereika, Ph.D.
School of Nursing
University of Pittsburgh
Pittsburgh, Pennsylvania

>Abstract The generalizability of research findings to clinical areas has traditionally been a measure of research value. This age of technology has provided a tool to further expand the use of research to rural areas in a cost-effective manner. Computers are being used not only for data collection and analysis, but also for clinical purposes such as health screening. This presentation will discuss the use of computers in rural schools for health screening. A University of Pittsburgh research team traveled distances of up to 2 hours to reach rural high schools to screen students for health problems. This screening process involved contact with 445 rural students who completed questionnaires during school. The 10 questionnaires provided information on drug use, self-esteem, coping, social support, depressive symptomatology, anger, optimism, and perceived physical health. Data entry, verification, and analysis were completed using Teleform for Windows, version 4.0, which uses optical character recognition of data input via scanner, the Paradox Database Management Program, and SPSS (Statistical Package for the Social Sciences) for Windows, version 6.1, at the University. Within 2 days of testing, students at risk for depression or self-harm were interviewed by Advanced Practice Nurses, then evaluated and referred for treatment. Students who were at moderate risk due to poor coping skills were asked to participate in a 10-week cognitive-behavioral group. This prevention program, Teaching Kids to Cope (TKC), not only targets depressive symptomatology, but also improves the repertoire of coping skills that may be effective for combating problem drinking and/or other substance abuse. The screening process used provided research-based identification of students at risk and direction for intensive professional intervention. The schools involved in this study are considering the integration of this screening process into the regular curriculum in the same manner as they now do vision and hearing screening.

In this climate of health care change, a major focus is prevention of illness, rather than one of healing after an illness has occurred. Adolescence is not usually thought of as a time fraught with major health concerns. However, the Office of Technology Assessment (1991) has examined recently available data which indicate that this may not be the case. According to this new information, many health problems, both physical and emotional, plague the adolescent population. Preventive health care for the adolescent has only recently been considered. It is of vital importance to evaluate students and target the "at risk" adolescent in order to offer suggestions for successful preventive intervention (Crockett and Petersen 1993).

Most of the recent studies concerning adolescent health have centered around urban teens, but youth in rural areas must also be carefully considered, researched, and evaluated. Preventive strategies must also appraise the social context.

One important developmental characteristic to take into account is the teen's current environment. Family and peer groups have a strong impact on the adolescent's environment. In rural areas, relationships may be more limited and thus exert a greater influence. This concentration of relationships occurs mainly because of the low population density and does have an impact on and make rural concerns unique.

Some rural adolescent issues are problems with access to care, often poverty with resultant inability to obtain health insurance, and a decrease in the availability of program funding (OTA 1990, 1991). Further research is needed to amend the already accumulated data on adolescent health issues.

Research should include plans for ease of identification and rapid intervention with rural teens who may be considered at a high level of risk. High-risk teens include those whose background would strongly indicate development of high-risk behaviors or actions. These behaviors may include, but are not limited to, promiscuity, violence, suicide attempts/completion, and substance abuse. Prevention is the key here, as it is easier to prevent the development of such behavior than it is to "heal" it.

Priorities for research, then, lie in the areas of health promotion with a strong focus on the high-risk populations of rural adolescents. Therefore, longitudinal assessment of adolescent health and the development of effective interventions to promote adolescent health and to enhance coping are public health priorities.

Purpose/Significance

The two goals of this longitudinal study are: (1) to measure mental and physical health in a sample of rural adolescents using computerized data entry and analysis, and (2) to test the effectiveness of a nurse-managed intervention designed to increase self-esteem, coping, and social support of students. Information obtained from the questionnaires used in the screening addressed adolescent substance use/abuse. If possible, correlations will be made, thus allowing for the development of improved preventive interventions.

A secondary objective of this study concerned the use of the Teleform data entry program. Use of this program allowed for rapid and accurate data entry. Our research team members were then able to return to the schools and assess any student who scored above 77 on the Reynolds Adolescent Depression Scale (RADS). These assessments were made within 2 days after testing. The combination of the depression score and the psychiatric assessment determined who was referred for followup care or eligible for participation in the randomized trial of the coping intervention.

The long-term significance of this study is the expectation to decrease the negative risk taking behaviors of adolescents and promote positive coping strategies. A conservative assessment of children's mental health (Pothier 1988) concluded that at least 12 percent of children, or 7.5 million children, were in need of some sort of mental health service. Nearly half of the children needing mental health service are deemed severely disordered or handicapped by their mental illness. The prevalence of child and adolescent mental illness may now range between 17 percent and 22 percent, or from 11 to 14 million children (NIMH 1990).

Teenage substance abusers have also increased in numbers, along with children who are at risk for mental illness because of environmental factors such as poverty, inadequate care, parental illness, death, and divorce. In addition to diagnosable mental disorders, statistics show that nearly 5,000 young people between the ages of 15 and 24 killed themselves in 1987 and that more than a million people in that age group make suicide attempts each year (CDC 1987).

Depressive symptoms, associated with or without suicidal ideation, may represent a potential life threatening crisis for the young person and indicate a need for prevention, evaluation, and intervention. According to the U.S. Department of Health and Human Services (1991), two of the major health objectives for the year 2000 are to decrease the adverse effects of stress by 18 percent and decrease the rate of suicides by 10 percent in the total population. Priorities for research, then, lie in areas of health promotion and disease prevention focusing on the high-risk populations of adolescents. One of the key areas to be explored is how to promote mental health, and to provide interventions for rural adolescents with depressive symptomatology by increasing social support and coping skills through education.

Specific Aims of the Study

The following are the specific aims of the study:

1. To describe the mental health of rural adolescents by identifying depressive symptomatology, perceived social support, coping strategies, drug use, self-esteem, anger, and optimism.
2. To describe the physical health of rural adolescents by identifying perceived physical functioning, bodily pain, vitality, general health, health needs and concerns, and illness-related absenteeism.
3. To examine the association of life events with mental and physical health as measured at baseline.
4. To evaluate the effect of the Teaching Kids to Cope (TKC) intervention on the mental health of rural adolescents.
5. To evaluate the effect of the Teaching Kids to Cope (TKC) intervention on the physical health of rural adolescents.

Program

The theoretical basis for this study is adolescent mental and physical health promotion. See figure 1. The 10-week Teaching Kids to Cope protocol is described briefly in table 1 and is described more fully in The Adolescent Mental Health Program Manual.

Adapted from Puskar et al. 1990. Reprinted with permission, Nursecom, Inc.

Methods

In Phase 1 of the study, which lasted from 1995 to 1996, 445 high school students were surveyed to establish baseline data on the identified parameters related to mental health (depressive symptomatology, social support, coping, drug use, self-esteem, anger, and optimism) life events and physical health as well as to identify intervention and control groups (students who score 66 and above on the Reynolds Adolescent Depression Scale [RADS]). Students who scored in the high range (above 77) on the RADS, or who responded positively to critical items indicating possible suicidal ideation, were interviewed immediately by members of the research team in collaboration with Services to Adolescents at Risk (STAR) from Western Psychiatric Institute and Clinic, Pittsburgh, Pennsylvania. The students were referred for treatment if clinical depression or suicide risk was determined. Data from these subjects will be examined separately.

The research sample included freshman, sophomore, and junior high school students from three schools located in rural settings of southwestern Pennsylvania. Each one of these high schools serves a small rural community of approximately 7,000 people. Average family income of each of the three areas is approximately $24,000. Minorities (5 percent African Americans) and both genders (50 percent male and 50 percent female) are represented. Dropout rates for the schools are 1.1 percent to 2 percent.

Inclusionary criteria included being enrolled in the regular, college preparatory, and honor classes; and ability to read and write English. Exclusionary criteria included being enrolled in a classroom designated for socially or emotionally disturbed children, and a loss through death of a parent or caregiver within 12 months prior to the study. It is believed that such a loss could increase depressive symptomatology and does not represent the usual high school student's life stress. Data from students identified by the school as learning disabled were deleted. Only two subjects were eliminated from the pilots based on being in learning disabled classes.

Students were asked to volunteer, and both their consent and that of their parents were obtained prior to testing or intervention. Mental and physical health and life events for all subjects were assessed at baseline. Based on the results of those surveys, the intervention was implemented and longitudinal assessment of intervention efficacy will be done in the future. For intervention/control groups, measures will be done at post intervention (T1), 6 months post intervention (T2), and 1 year post intervention (T3). Teacher ratings of intervention/control subjects' behavior will be done at baseline, T1, T2, and T3. At each assessment point, subjects' school records will be evaluated for grades, absences, and any discipline reports. The intervention group will receive the Teaching Kids to Cope intervention, a 10-week group protocol that has been shown to be effective in decreasing depressive symptoms and increasing coping skills in adolescents. A "booster" session for the groups will be given 9 months following the intervention. The booster should help to determine if such a session can enhance long-term effects of the intervention. This study builds on preliminary work evaluating access, retention, and efficacy of the intervention protocol. It further tests the protocol in promoting coping with a larger sample, evaluates the effect of a booster session, and assesses changes over time. In addition to self-reports, which are often not as reliable, student behavior will be evaluated by teacher reports and school records.

At-risk students can be identified quickly when computerized data entry systems such as Teleform are used. All data collection instruments used in the study were generated using Teleform for Windows, version 4.0 (Cardiff Software, San Marcos, California), which uses optical recognition of data input via scanner. The instruments used were adapted using the designer module of Teleform for Windows. The Teleform Designer uses text object and data entry fields. Several data entry fields are available, including choice, entry, constrained print, and image. Choice and entry fields allow the respondent to select a response by shading a bubble next to one of several listed options. Image fields are used to capture text data, which are generally of greater length. Text responses to open ended questions are captured using the constrained print or image fields.

As a rule, choice or entry fields were used whenever possible due to ease of use, high degree of accuracy, and the decreased need for visual verification. Constrained print fields were used whenever open ended questions requiring short responses (eight characters or less) were used. Constrained print fields are subject to greater interpretation by the computer and as a result were always visually inspected to ensure accuracy. Image fields were used when longer response, were needed (on the Child Behavior Checklist—Youth Version).

Finalized forms were then reproduced and distributed to subjects. Completed forms were scanned using the Teleform Reader and a Hewlett Packard Scanjet III. Once scanned, the images of the questionnaires were evaluated and held for verification. Images were verified using the Teleform Verifier. This program displayed the image and moved through each questionnaire, allowing for visual verification of each constrained print field and image field, as well as any other type of field that was not answered. When the computer stopped, it required that the person entering the data look at the field on the computer screen and check the question against the actual questionnaire. Changes could be made according to the response of the student. Fields that were not interpreted correctly by the computer were corrected. This process was conducted by the Project Data Management Specialist and Director/Research Associate. Several computerized options are available when using Teleform to increase the accuracy and quality of the data. One such option employed was to make every field "entry required." This option caused the program to stop at every field/question that was left blank, regardless of its type.

The corrected questionnaires were then transferred by the Teleform Verifier to the Paradox Database Management Program. Scoring programs were generated using Paradox. Demographic characteristics and summarized scores were transferred to SPSS for Windows for appropriate statistical analyses.

The amount of data to be obtained from the survey is considerable. The optical character recognition program that we used provided rapid and accurate data entry and reduced turnaround time for interventions. In our study, the use of optical character recognition proved to be very effective, as the research team saw any student determined to be at risk in terms of their RADS score within 2 days. The only time this was not possible was if the questionnaires were administered on a Friday, in which case the students were seen on Monday. Students who were absent when they would ordinarily have been seen were seen as soon as possible after their return to school.

With this type of program, a data entry system similar to ours could be used by school or health professionals when surveying rural adolescent populations. The school personnel would administer the surveys and use a facsimile system to transfer the results to the computer in a university setting. Following analysis of these data, the rural clinicians could focus their interventions appropriately. This would result in maximum cost effectiveness and positive student outcomes. It would also, then, reduce the distance between university expertise and the at-risk rural adolescent.

Research Design

The research is being conducted in two phases, across three different class years within three different rural high schools. Phase 1 consisted of a survey to collect data on all 9th, 10th, and 11th grade students in each high school who were able to participate. The psychological battery used for the survey was comprised of 10 instruments: Reynolds Adolescent Depression Scale, Child Behavior Check List, Anger Expression Scale, Adolescent Health Inventory, Drug Use Screening Inventory, Life Orientation Test, Coping Responses Inventory Youth Form, Perceived Social Support Scale, Life Events Checklist, and the self-report for childhood anxiety related disorders (SCAReD). The physical health component involved use of the Adolescent Health Inventory. These survey data will provide information that can be used to describe adolescents in general (specific aim no. 1, no. 2) and examine the interrelationships among stressful events, and outcome measures (specific aim no. 3) as well as serve as baseline data (specific aim no. 4, no. 5) for those eligible students who are later randomized to a treatment protocol.

Phase 2 of the study, which will extend from 1996 to 1997, will utilize an experimental, two-group design to examine the efficacy of Teaching Kids to Cope on optimism, self-esteem, perceived social support, variety of coping strategies used, anger, drug use, physical health, and depressive symptomatology for students who scored in the moderate to high range (66 and above) on the RADS. Based on the results of phase 1, subjects surveyed in phase 1 who meet the criterion of 66 or above on the RADS, are not at suicidal risk as determined by interview, and agree to participate, will be randomly assigned to either the control or intervention group. The 10-week psycho-educational TKC groups that meet during school hours are currently being conducted in the three schools. Each of the three schools has a control group and an intervention group, consisting of students from the 9th, 10th, and 11th grades.

Randomization to treatment protocol will be blocked, with blocks being based on the school (A, B, and C), class year (9th, 10th, and 11th), and gender (females and males), resulting in 332=18 school/class/ gender randomization blocks. By randomizing to treatment within school/class/gender blocks, imbalances as to the number of subjects in each of the treatment groups across the combinations of school, class year, and gender are minimized. As to the sequencing of treatment assignment with the school/class blocks, treatment blocks of size 4 were used, resulting in six possible combinations of the two treatments across four positions in the treatment block.

The assessment measures will be the same as those used in the Phase I survey:

• Depression (RADS)
• Child behavior (Child Behavior Check List)
• Teacher report (Child Behavior Check List)
• Anger (Anger Expression Scale)
• Drug use (Drug Use Screening Inventory)
• Optimism (Life Orientation Test)
• Coping (Coping Responses Inventory—Youth Form)
• Social support (Perceived Social Support Scale)
• Life events (Life Events Check List)
• School records
• Physical health
• Anxiety (SCAReD)

Using these measures, randomized intervention subjects will be reevaluated immediately following the conclusion of the intervention (Time1), 6 months post intervention (Time2), and 1 year post intervention (Time3). Control subjects will also be reevaluated using the survey instruments at three points in time but at 10 weeks post randomization (Time1), 6 months post Time1 (Time2), and 1 year post Time1 (Time3). Teacher behavior ratings and school records (grades, absences, disciplinary reports) will also be assessed at baseline, T1, T2 and T3 for intervention and control subjects. A "booster" session will be given 9 months following the intervention to a random sample of half of the intervention group for each school/class block. This will be done to see if a booster session can enhance long-term effects of the intervention. During the 45-minute booster session the group leader will review different coping methods and cognitive distortions and discuss how students have applied their knowledge about cognitive distortions through sharing of examples since the conclusion of TKC.

Barriers/Problems Encountered and Solutions

Difficulties were encountered when researchers left the University of Pittsburgh and entered the high school educational domain. Much effort was expended in compromise and good will to elicit the cooperation of school personnel.

Obtaining consent forms was problematic. It required frequent contact with the student populations and encouragement of the teachers to help in their return. The students were assured of confidentiality, and it is hoped that this will allow for truthful self-reports.

Processing the data continues to be a tedious procedure. All the forms must be viewed to ensure accuracy of scanning. It was extremely difficult to have the students print and shade correctly. However, University staff are currently revising and streamlining this process.

Current Status of the Study

The research team has screened 445 rural high school students with the psychological battery of 10 instruments. Of these students, 10.3 percent, or 46 students, had RADS above 77 and have received individual assessments.

Of the students interviewed, 43 percent were referred for treatment. It is unknown what percentage will follow through with treatment. The lack of available treatment resources as well as student reluctance to attend treatment is an ongoing issue in rural school settings.

The second intervention phase of the study is a 10-week psycho-educational group that meets during school hours. The groups are currently being conducted in the three schools. Following completion of the study, data will be analyzed and conclusions reached.

The early evidence suggests the need for mental health screening in the schools and also for coping skills to become an integrated aspect of the curriculum. It may be possible through the use of a data entry system for scoring and coping skills groups during school hours that adolescents may be saved years of anguish and emotional suffering.

References

Beck, A. Cognitive Therapy and the Emotional Disorders. New York: New American Library, 1976.

Centers for Disease Control. Youth suicide: United States 1970-1980. Morbidity and Mortality Weekly Report 36(6):87-89, 1987.

Crockett, L.J., and Petersen, A.C. Adolescent development: Health risks and opportunities for health promotion. In: Millstein, S.G.; Petersen, A.C.; and Nightingale, E.O., eds. Promoting the Health of Adolescents: New Directions for the Twenty-first Century. New York: Oxford University Press, 1993.

Freeman, A., and Greenwood, V., eds. Cognitive Therapy: Applications in Psychiatric and Medical Settings. New York: Plenum Press, 1987.

National Institute of Mental Health. National Plan for Research on Child and Adolescent Mental Disorders. DHHS Publication No. (ADM) 90-1683. Rockville, MD: National Institute of Mental Health, 1990.

Office of Technology Assessment, U.S. Congress. Health Care in Rural America. OTA-H-434. Washington, DC: U.S. Govt. Print. Off., 1990.

Office of Technology Assessment, U.S. Congress. Adolescent Health. Volume III. Crosscutting Issues in the Delivery of Health and Related Services. OTA-H-467. Washington DC: U.S. Govt. Print. Off., 1991.

Pothier, P.C. Child mental health problems and policy. Archives of Psychiatric Nursing 2(3):165-169, 1988.

Puskar, K.; Lamb, J.; and Martsoff, D. Role of the psychiatric nurse in an adolescent coping skills group. Journal of Child and Adolescent Psychiatric Nursing 3(2):47-51, 1990.

Sank, L., and Schaffer, C. A Therapist's Manual for Cognitive Behavior Therapy in Groups. New York: Plenum Press, 1984.

U.S. Department of Health and Human Services, Public Health Service. Healthy People 2000: National Health Promotion and Disease Prevention Objectives. DHHS Pub. No. (PHS) 91-50213. Washington, DC: U.S. Govt. Print. Off., 1991.

Yalom, I. The Theory and Practice of Group Psychotherapy. 3d ed. New York: Basic Books, 1985.

Last Updated 11-7-02