- Email: [email protected]
School-based clinic use and school performance
JOURNAL OF ADOLESCENT HEALTH 1993;14:91-98
MARCELLA
T. McCORD,
JANE M. FOY, M.D.,
M.D.,
M.P.H.,
JONATHAN
AND KATE FOTHERGILL,
School-based health clinics (SBCs) have been promoted as an innovative approach to providing adolescent health care. The present study examined the effect of a SBC on academic success. We studied the effect of clinic registration and use on students’ absence, suspension, withdrawal, and graduation or promotion rates in an alternative high school for students who were not able to succeed in traditional educational programs. On average, these 322 high-risk students attended school only 58% of the time; 24% were suspended; and only 26% graduated or were promoted. The 189 (59%) students who were registered to use the clinic and the 159 (49%) who actually used the clinic were as likely to be absent or to be suspended as non-registered students. However, students who used the clinic were significantly more likely to stay in school, and to graduate or be promoted than students who were not registered for the clinic. This relationship was strongest for black males; those who used the clinic were nearly three times [odds ratio = 0.35; 95% confidence interval (CI) = (0.16-0.78~1more likely to stay in school than those who did not use the clinic. In multiple linear regression models predicting school performance, only clinic use and percent of enrolled days absent were significantly associated with graduation/promotion, and these two variables predicted 23% of the variance in promotion status.
From the Departments of Maternaland Child Health (M.T.M.) and Health Behavior and Health Education (RF.), the School of Public Health, and The Cecil G. Sheps Center for He&h ServicesResearch (I.D.K.), The t,tniversity of North Carolina at Chapel Hill; the Guilford County Health Department, Child Health Division, Greensboro, North Carolina (J.M.F.) and the Division of AdolescentMedicine, Department of Pediatrics, Universitrr of Rochester School of Medicine, Rochester, New York (J.D.K.1. Address reprint requests to: ]onathan D. Klein, M.D., M.P.H., Department of Pediatrics, Division of Adolescent Medicine, Box 690, Dniversity of RochesterSchoolof Medicine, 601 Elmwood Avenue, Rockester, Neza York 14642. Manuscript accepted October 7‘1992.
0 Society for Adolescent Medicine, 1993 P&4ished by Elsevier Science Publishing Co., Inc., 655 Avenue of
D. KLEIN,
M.D.,
M.P.H.,
M.P.H.
KEY WORDS:
School-based
clinic
School performance Adolescent School dropout
The school dropout rate for adolescents in the United States in 1989 was 12.6% (l), and that for North Carolina in 1990-1991 was 5.6% (2). Adolescents drop out for a variety of reasons: boredom or academic failure, health problems (i.e., mental health crises, substance abuse, pregnancy), family concerns, financial considerations, or other problems (1). The transition to high school has been associated with decline in grade point average (3,4), and poorer school attendance (4); this may have long-term implications for adolescents’ adjustment (4), as low grades and poor attendance correlate highly with future school dropout (5-7). Adolescents who do not complete high school are at increased risk for multiple health, economic, and social problems including unemployment, welfare dependency, and homelessness (1). Schools are the only institution regularly attended by most young people ages 5 through 16 years; nearly 95% of all children and youth are in elementary or secondzry schools (8). An effective strategy for improving the health behavior of society is to focus on youth as they develop health habits and to provide adolescents with the knowledge, skills, and motivation for engaging in healthy behaviors (8). Thus, schools represent a public institution with great opportunity for improving the health of the school-age population (8). Comprehensive schoolbased health programs, including school-based health clinics (SBCs), have been successful at im91
the Americas, New York, NY 10010
92
McCORD ET AL.
proving adolescents’ access to health and related services (1,g); increasing utilization of health care; familiarizing students with the health-care system; and identifying and treating adolescents’ health problems (9). There are now more than 300 SBCs operating in middle schools and high schools across the United States(10). Many of the students served by these clinics are low-income minority youth, with limited access to other sources of care (11). SBCs are well used by students in the schools they serve, with utilization rates typically reaching over 75% of inschool adolescents (11,12). For about half of enrolled students, school-based clinics are their sole or primary source of health care (1l), suggesting that these services and sites are acceptable to adolescents (12). Most published articles on SBCs have either been descriptive (13-16) or have taken an advocacy position without empirical data (17-21). However, some programs have reported success in improving contraceptive use, decreasing pregnancy and substance use rates, preventing school dropout, and improving school attendance (22-25). Two studies report SBCs improved attendance and lowered dropout rates (15,25). One study of pregnant and parenting adolescents in four cities showed a trend toward higher graduation rates (25), and the American School Health Association has presented testimony showing that 38% of the students in one N.Y.C. school reported that their attendance had improved as a result of the clinic (22). Dryfoos (15), however, has noted that school principals and advocates often believe SBCs positively effect absenteeism and dropout, these claims have not been supported by published data. In 1987, the Gillespie Park Education Center (subsequently renamed the Alternative Education Program or AEP), an alternative public high school and middle school for Greensboro, North Carolina Public Schools’ students whose needs cannot be met in a traditional educational program, received a Robert Wood Johnson Foundation grants to develop a SBC (26). Our goal in this study was to identify scholastic outcomes associated with use of the GHC; specifitally, to evaluate whether or not clinic use had an effect on students’ absences, suspensions, dropping Out, and graduation or promotion. We hypothesized that students who used the school-based clinic would have fewer absences and suspensions, would be less likely to drop out of school, and would be more likely to be promoted or graduate.
JOURNAL 03 ADOLESCZNT HEALTH Vol. 14, No. 2
Methods Sample and Site The AEP serves grades 6 through 12; 78% of the students are rnzorities, and most are economically disadvantaged (27). The student population at the AEP is a high-risk population, having failed to succeed in traditional schools. More than half of the students live in single-parent households headed by females. Approximately 22% of the students are selfsupporting, and approximately 4% live independently from their parents. Over 30% of the student population are either pregnant or parenting a child. In 1986, 35% of the AEP students reported having used “hard” drugs (other than alcohol, cigarettes, and marijuana) in the past year. Chronic attendance problems, academic underachievem.ent, and serious behavior problems are frequent among the AEP students. In a 1986 survey, the AEP students’ health risks were significantly higher than those of other Guilford County students in 13 of 15 categories tested, including diet, dental hygiene, exercise, smoking, use of alcohol and drugs, unsafe driving practices, and stress (27). Before the Gillespie Student Health Project (GSHP) opened, more than 25% of the AEP families surveyed cited transportation and distance as obstacles to obtaining health care (27). Almost half also reported not being able to afford health care. The grant funds the GSHP, which consists of health education, health promotion, screening, and clinical services, all delivered via the Gillespie Health Center (GHC). The all-female staff of the clinic consists of a part-time physician, nurse practitioner, and clerk, and a full-time registered nurse, socia! worker, and clerk. The project is administered jointly by the Guilford County Department of Public Health, Moses Cone Hospital, and the Greensboro Public Schools. Data Collection We collected archival data from the AEP, GHC, and the Guilford County Health Department about school enrollment, attendance, suspension, drop out, and graduation/promotion for all students who attended the AEP during the 1990-1991 academic year. This was the clinic’s fourth year of operation. School and demographic data were abstracted from school attendance records. Clinic registration and utilization data, as well as additional demographic information, were obtained from the clinic. For an adolescent to be registered in the clinic, his or her
March 1993
SCHOOL-BASED CLINIC USE AND SCHOOL PERFORMANCE
parent must first have a face-to-face interview with clinic staff, either at school or during a home visit. Files were merged by name and confidentiality was maintained by removal of all student identifiers. The data were obtained through ongoing program evaIuation and confidential data-sharing agreements between the school and the Guilford County Health Department. Data were entered using Epi Info, Version 3 (28) directly into a portable computer from school, clinic and health department files.. Statistical Analysis
All analyses were conducted using the SAS statistical analysis system, version 6.0 (29) for personal computers. The statistical significance of observed differences between group mean values was done with the student’s t test. Analysis of variance was done using SAS general linear regression modeling to compare the mean values across multiple groups. This procedure was also used for multiple linear regression modeling to measure associations between the school-performance variables, clinic use, and demographic factors. For bivariate analyses, x2 tests were used to determine whether observed differences in proportions between study groups were statistically significant. Fisher’s exact test was used to compare differences between white student’s suspensions and promotions, as these cell sizes were quite small. Statistical significance for stratified data was determined using the Mantel-Haenszel test with one degree of freedom (30). In all cases, a p value of 0.05 was considered statistically significant.
Results During the academic year of August 1990-May 1991, 322 students were enrolled at the AEP. Mean time of enrollment was 102 days out of the 180-day academic year. Fifty-two percent of the AEP students were female, and 85% were black. The 189 students who were registered to use the GHC were similar to all enrolled students by gender, ethnic@, and age (Table 1). The population was also similar with respect to grade in school; however, because this is an alternative school, grade has little meaning. Because of this, age is used in the regression analyses. School Absence, Suspension, Promotion Status
and
Students were absent (including excused absences, unexcused absences, and suspensions) an average
93
Table 1. Summary of Enrolled Populationand Registered Population 1990-1991 Academic S&OO~ Year
Gender Female Male Ethnicity Black White Gender and ethnic@ Black female Black male White female White male Age (years) 12-13 14 15 16 17 18 19-20
School enrollees (n = 322) (B)
Cliic enrollees (n = 189’ (%) __
52 48
54 46
85 13
84 16
47 40 6 7
48 36 8 9
12 15 25 18 15 10 52
17 20 30 11 13 6 4
of 33.7 days. Excluding suspensions, AEP students were absent an average of 31.2 days. Nearly 20% of students were absent 50 or more days. When absences were considered as a proportion of the number of days enrolled in school, female students were absent a higher percent of time than males, and whites were absent a higher percent of time than blacks (Table 2). Overall, 78 (24%) students were suspended during the 1990-1991 school year. Among students who were suspended, the average number of days of suspension -was 2.5 days. Six students were suspended for between 21 and 40 days and one student was
suspended for 57 days. More black students were suspended than whites; and black students, on average, were suspended for a higher percentage of the time they were enrolled in school than were white students. Male students also were suspended a higher percentage of the time they were enrolled than were female students (Table 2). Black males were suspended an average of 6% of the time they were enrulled, compared with 1% for black females, white males, or white females. Eighty-four (26%) students were promoted or graduated during the 1990-1991 school year: 8 (3%) students graduated, 50 (16%) were promoted at the end of the year, and 26 (8%) were promoted midyear.
Nearly two-thirds (62%) of all students with-
94
JOURNAL OF ADOLESCENT HEALTH Vol. 14, No. 2
l&CORD ET AL.
Table 2. Average Percent of Time Absent and Suspended While Enrolled and Average Number of Clinic Visits
Time absent (%) Time suspended (96) Averape number of clinic visits
Female (n = 145)
Male (n = 123)
p value
Black (n = 226)
White (n = 37)
p value
42 1 6
35 5 4
0.005 0.0001 0.004
38 3 5
49 1 4
0.005 0.002 0.69
drew from school at least once during the academic year, Of these 43 (21.4%) m-entered school and 32
(74%) withdrew from school a second time. Nearly three-fourths of AEP students were retained in their grade in 1990-1991. On average, students who graduated or were promoted had fewer absences or suspensions as a percentage of their time enrolled in school than those students who were retained (Table 2). Males, females, blacks, and whites were all equally likely to graduate or be promoted.
Clinic Registration and Visits A total of 14!57visits were made to the clinic during the 1990-1991 school year. In addition to the 322 students enrolled at the AEP, 27 ex-students (dropouts, graduates, transfers to other schools) and 47 children of the AEP students (or ex-students) used the clinic. The students who were enrolled in the ABP for the 19!&1991 school year made 1060 visits.
Of these 322 students, 189 (59%) had registered to use the clinic, and 159 of these actually used the clinic one or more times. The average number of clinic visits for the 189 registered students was 4.65. Of the 189 registered students, 30 (16%) made no visits. About half (55.0%) of clinic re-:strants made one health-screening visit, and half (50%) of the registered students also made between c s?t and five treatment visits, 15% made 6-10 treat.nr;lt visits. Blacks were nearly twice as likely not to have registered as whites (44% versus 27%; p < O&Q, however, once registered, blacks and whites made the same average number of clinic visits (Table 2). Males and females were equally likely to be registered for the clinic. On average, however, females made more clinic visits than males (Table 2). Clinic Use and School Performance Overall, students who registered for the clinic were enrolled in school an average of 112 days. Those who were not registered to use the clinic were enrolled an average of 92 days (p < 0.002). The percent of time students were absent while they were enrolled did not differ for students registered and not
Retained Promoted (n = 186) (n = 82) 44 4 3
26 2 8
.-
p value o.OOBO1 0.025 o.OOB1
registered for the clinic (38% versus 41%; p = 0.19). However, students who actually used the clinic were absent a lower percent of the time they were enrolled in school (36%), compared to students who registered but did not use the clinic (51%) and students who did not register (41%) (p C 0.02). More students who were registered for the clinic were suspended than students who were not registered (30% versus 16%) (x2 = 8.781, p < 0.003).
Controlling for race and gender (Table 3), we found that black students who had registered were twice as likely to be suspended than those not registered, and black males were nearly three times as likely to be suspended if registered for the clinic. Black male and black female students who registered and used the clinic were even more likely to be suspended than those who registered but never made a visit. However, there is no difference between the actual percent of time spent suspended while enrolled in school for registered students and students who registered but did not make a visit. Students who were registered for the clinic were more likely to stay in school than students who were
Table 3. Adjusted Odds Ratios Comparing Registered
and Non-RegisteredStudent Odds of Withdrawal from School, BeingSuspended, and Being Promoted (By Race and Gender) Withdrawal
All0 Blacks Whites Females Males Black females
Black males White females White males
0.5** 0 4** 1:9 0.6 0.4* 0.5* 0 3** 1:2 2.2
Suspension 2.3” 2 3** 0:o 2.7, 2.4* 2.4 2.8’+ 0.0 0.0
Promotion 1.8* 2.1” 0.7 2.0 1.5 2.3’ 1.7 0.5 0.8
“Referencecategory including all students at the school (i.e., for all students, those who were registered were half as likely to withdraw from school as those not registered). p value < 0.05. l*p value < 0.01. N.B.: The associations between the study variables vanes slightlv by the different strata of the control variables, however, this variance is not supported by a significant Breslow-Day test for any of the variables.
SCHOOL-BASED CLINIC USE AND SCHOOL PERFORMANCE
March 1993
Table 4. Percentages of Students Who Stayed in School and Were Promoted Studentswho stayed in school (%) Female (168) Male (154) Black (274) White (41) Total (322) Students
promoted
Female (168) Male (154) Black (274) White (41) Total (322)
Registered Not registered for clinic
for clinic
p value
40 49 48 23 44
28 29 28 36 29
0.10 0.02 0.001 0.044 0.005
36 24 33 20 31
22 18 19 27 20
0.05 0.30 0.00 0.68 0.02
(a)
not registered (44% versus 29%) (x2 = 7.835, p < 0.005). Almost half (48%) of black students registered for the clinic stayed in school compared to 28% of non-registered black students (Table 4). Registered male students were also twice as likely to stay in school than non-registered students. Controlling for both race and gender (Table 3), black males were nearly three times as likely to stay in school if they had registered for the clinic. Black female students who were registered were also more likely to stay in school than those not registered. Among whites, registered students were no more likely to stay in school than those not registered. When we examined the likelihood of staying in school by whether students actually used the clinic services, we found that students who registered and used the clinic were more likely to stay in school than those who registered but never made a visit. Almost half (48%) of those students who used the clinic (compared to only 23% of registered students
95
who did not make clinic visits) stayed in school. This relationship was striking among black males; 60% of those who registered and used the clinic stayed in school compared to only 18% of those who were registered but did not use the clinic (Figure 1). Students who were registered were also more likely to graduate or be promoted than students who were not registered (31% versus 20%) (x2 = 4.86, p < 0.03). Students who were registered were nearly twice as likely to graduate or be promoted [odds ratio (OR) = 1.81; 95% confidence interval (CI) = (1.067-3.054)]. Students who graduated or were promoted averaged eight clinic visits compared to three visits made by students who were retained in their grade (p < 0.0001) (Table 2). Among blacks, 33% of students who registered for the clinic graduated or were promoted compared to 19% of non-registered students (Table 4). Among whites, registered students were no more likely to graduate or be promoted than those who were not registered. Students who registered and actually used the clinic services were more likely to graduate or be promoted than those who registered but never made a visit. Nearly two-thirds of all black males who graduated or were promoted had used the clinic (Figure 2).
uItivariate
Linear Regression
We used multiple linear regression models (SAS GLM procedure) to assess the strength of the association among clinic visits, and absences and promotion or graduation status. In these analyses, the total number of clinic visits was the independent variable.
60
Percentwho my6d in school
40
who W.R promoted
Black Female
Figure 1. Percent of clinic registrants, users, and non-users who stayed in school by race/gender groups. Overall p = 0.077 for black females, p = 0.001 for black males, p = 0.857for white females, and p = 0.592 @ white males.
Figure 2. Percentage of clinic registrants, users, and non-users who graduated or were promoted by race/gender groups. Qoerall p = 0.034 for blackfemales and p = 0.034 for black nudes. Too fewwhite students were graduated to de$ct in this figure.
96
JOURNAL OF ADOLESCENTHEALTH Vol. 14, No. 2
McCORD ET AL.
The independent variables did not determine very much of the variance for percent of time absent (tz = 0.083) or percent of time suspended (? = 0.099). However, the model accounted for a substantial portion of the variance in student’s promotion status (J = 0.233) (Table 5). In a stepwise model, the only two significant predictors of promotion status were percent of time absent (partial 3 = 0.180; p = 0.0001) and total number of clinic visits (partial ? = 0.047; p = 0.0001). Chances of promotion were increased by fewer absences and a higher number of clinic visits.
Discussion We expected that students involved with the schoolbased clinic at the AEP would have fewer absences or suspensions, would be more likely to stay in school, and would be more likely to graduate or be promoted to the next grade. Our results partially supported our hypotheses. For example, we found no difference in the days absent or suspended while enrolled between students registered and not registered for the clinic. Stud.ents who used the clinic were only slightly less likely to be absent than nonusers. However, registered students and students who actually used the clinic were twice as likely to stay in school and nearly twice as likely to graduate or be promoted than non-registered students. Additionally, the greater students’ exposure to the clinic (i.e., actual visits to the clinic), the stronger all of the relationships between clinic use and graduation or promotion became. The relationship between clinic use and staying in school was particularly striking among the black males at Gillespie. Males, especially those in minority groups, are traditionally the most difficult group to reach with adolescent health services (1,ll). Black males also drop out of school at higher Table5. Regression Models and B Estimates for Student’s Absences, Suspensions, and Promotion Time absent Time suspended WI (%I Total 9 (F;p value) Age Gender Ehicity Clinic visits Percent absences Percent suspensions
(s.8llzL1, -0.704 -5.262 0.230 -1.265
(7.1ozLo1) -0.443 4.221 -3.280 NS
Promotion or graduation (r2.9gzIoo*) - 0.010 -0.068 - 0.013 0.020
-
-
-0.008
-
-
NS
than either black females or whites (1). In this population, black males who used the clinic were three times as likely to stay in school as black male clinic non-users, and two-thirds of black males who graduated or were promoted were clinic users. The entire student body at the AEP is at high risk for school failure. Thus, unfortunately, we were unable to study the way in which students who registered differed from those not registered. Part of the difference may lie in selection bias in the adolescents themselves. While this group may be more self-motivated and desire better access to health services, this might also have been expected to affect school attendance, which is similar for clinic registrants and non-registrants. Another possibility is that these students may have more health problems and therefore may need clinic services more than others. If health status among clinic users is worse than that of non-registrants, this might help explain absenteeism being similar and not lower among the clinic population. A third factor may be attributable to the adolescents’ parents. For an adolescent to be registered in the clinic, his or her parents must have had a faceto-face interview with clinic staff, either at school or at a home visit. Thus, clinic registration may reflect the parents’ involvement in their child’s education. A final factor may be attributable to the school staff. School authorities noted that the school and clinic staff may, in fact, try hardest to reach students at highest risk. For both males and females, the students at highest risk in an already high-risk population may be targeted to receive the greatest number of services. Students who used the clinic averaged a slightly lower percent of time absent than non-users and ‘or the clinic. Ninetythose students not register four percent of the AEP shtdents were absent for one day or more, and 70% were absent more than 14 days. In other North Carolina schools, only 9% of students are absent for more than 14 days (2). In Guilford County, this figure is 8% (2). In other innercity school, estimates of daily school absence rates are between 20% and 40% (7,31). The students at the AEP were absent slightly over two-fifths of the total amount of time they were enrolled. Another limitation of this study is that we were unable to differentiate between absenteeism for “health” reasons or for other reasons. This is particularly true for pregnant adolescents who are “homebound,” who are counted absent at the AEP even though they are receiving home curricula. This may account for up to 200 days absent per term (32). Additionally, rates
March 1993
SCHOOL-BASED CLINIC USE AND SCHOOL PERFORMANCE
clinics cannot prevent many of the illnesses that might keep students home, and clinic staff may in fact facilitate absences by correctly diagnosing illnesses that should be treated by bed rest (11). These two factors might help explain the lack of difference in percentage of time absent between clinic users and non-users, at least among females. We are encouraged by the fact that the chance of staying in school is increased by students’ contact with the clinic. Overall, 62% of the population of students at the AEP enrolled for the 1990-2991 academic year withdrew from school. In North Carolina, 45% of persons 25 years or older in 1980 reported not having completed high school (33). The dropout rate for North Carolina for 1990-1991 was 5.6%, and for Guilford County it was 3% (2). In 1989, only 52% (56% of whites and 32% of blacks) of high school dropouts were employed, whereas 76% (79% of whites and 60% of blacks) of high school graduates were employed (34). Although the chances of adolescents at the AEP staying in school is low, students having greater contact with the clinic were more likely to stay in school. The reasons that these adolescents stay in school cannot be understood from this study. The AEP and the GHC aim’ to provide a supportive environment that will encourage teenagers to stay and finish high school. While use of the clinic is unlikely to be the actual cause of better school performance, a strong relationship exists between number of clinic visits and school promotion. A possible explanation is that the trust and support provided by SBC personnel helps students identify with and function better in school. We believe that SBCs provide positive role models and social support for the adolescents they serve. By having accessible, affordable, and confidential services at the school, students are provided not only with a convenient source of health care. They also may in fact gain a source of increased social support. Students may benefit from both the services rendered at the clinic and the perception that support services are available. Because we were limited to secondary data, we did not attempt to measure actual or perceived social support. However, this conceptual model is useful in helping to explain our findings. In general, excessive absenteeism pre diets those adolescents who are at highest risk for school drop-out and failure (35). In our population,
there is a school-wide problem with excessive absence, but some of these adolescents stay in school, and these are the students who have used the SBC. These students are not only staying in school, but
97
they are also succeeding academically, successfully moving ahead to higher grade levels, and fulfilling requirements for graduation. Understanding these phenomena will require prospective studies of students at schools with and without school-based clinics. This study was supported in part by the School-Based Health Program of The Robert Wood Johnson Foundation and by grant R48CCR402177 from the Centers for Disease Control. The authors thank Julia G. Lear and Elizabeth McAnamey for commenting on earlier drafts of this manuscript, Carol Porter for computer programming advice, and the staff of the Greensboro Public schools’ Alternative Education Program, the Gillespie Health Center and the Guilford County Department of Public Health for their assistance in data collection.
References 1. Office of Technology Assessment. Adolescent Health Volume 1. Summary and Policy Options. United States Congress, 1991. 2. The North Carolina Public Schools 1991 Statistical Profile. Annual publication. Raleigh, NC: N.C. Department of Public Instruction, 1991. 3. Blyth DD, Simmons RG, Carlton-Ford S. The adjustment of early adolescents to school transitions. Br Educ Res J 1983;8:149-64. 4. Felner RD, Faber SS, Primvera J. Transitions and stressful life events: A model for primary prevention. In: Felner RD, Jason LA, Moritsugu JN, et al, eds. Preventive Psychology: Theory, Research and Practice in Community Intervention. New York: Pergamon, 1983. 5. U.S. Department of Health, Education, and Welfare. Dropout Prevention. Washington, DC: Education Resources Information Center, 1975. 6. Weitzman M, Klerman LV, Alpert JJ, et al. Factors associated with excessive school absence. Pediatrician 1986;13:74-80. 7. Klerman LV, Weitzman M, Alpert JJ, et al. Why adolescents do not attend school: The views of students and parents. J Adolesc Health Care, 1987;8:425-30. 8. Kirby D. Comprehensive school health and the larger community: Issues and a possible scenario. J Sch Health 1990;60:170-7. 9. Council of Scientific Affairs. Providing medical services through school-based health programs. J Sch Health 1990;60:87-91. 10. Waszak C, Neidell S. School-Based and School-Linked Clinics: Update 1991. Washington, DC: Center for Population Options, 1991. 11. Kirby D, Waszak CW, Ziegler J. An Assessment of Six SchoolBased Clinics: Services, Impact and Potential. Washington, DC Center for Population Options, 1991. 12. Klein JD, Slap GB, Elster A, et al. Access to health care for adolescents: A position paper for the Society of Adolescent Medicine. J Adolesc Health Care 1992;13:162-70. 13. Blum RW, Pfaffinger K, Donald WB. A school-based comprehensive health clinic for adolescents. J Sch Health 1982;52:486-90. 14. Edwards LE. Steinman, ME, Hakanson, EY. An experimental comprehensive high school clinic. Am J Public Health 1977; 673765-6.
98
JOURNAL OF ADOLESCENTHEALTH Vol. 14, No. 2
McCORD ET AL.
15. hyfoos J. School-based health clinics: A new approach to y;nz;tinn5adolescent pregnancy? Fam Plann Perspect ;: . 16. Kirby D. Comprehensive school-based health clinics: A growing mcvement to improve adolescent health and reduce teenage pregnancy. J Sch Health 1986;56:289-91. 17. Cherskov M. SchooI clinics help kids and hospitals, too. Hospitals 1986;60:95. 18. Davis JM, Devaney, JM. Comprehensive centers. West J Med 1986;144:625-6.
school-based
teen
19. Gonzales C, Mu!liganD, Kaufman A. Adolescent health care: Improving access by school-based services. J Fam Pratt 1985;21:26%70. 20. Keenen T. School-based adolescent Pediatr Nurs 1966;12:365-9.
health care programs.
21. Lisoskie S. Health clinics in high schools. J Am Med Worn
Aasoc 19&5;41:106-126. 22. American School Health Association. Testimony on the Year 2ooo Objectives for the Nation: Recommendations for the School Health Program. Dayton, OH: April, 1988. 23. Dryfoos JG. School-based health clinics: Three years experience. Fam Plann Perspect 1988;20:193-200. 24. Joiner T. The role and success of school-based Ford Hasp Med J 1986;36:230-1.
clinics. Henry
25. Zabin LS, Hirsch MB, Smith EA, et al. Evaluation of a pregnancy-prevention program for urban teenagers. Fam Plann Perspect 1986;18:119-26.
26. Robert Wood Johnson Foundation. The School-Based Adolescent Health Care Program. Washington, DC: April, 1986. 27. Guilford County Health Department. Needs Assessment for Adolescent Health in Guilford County, 1985-86. Greensboro, NC. 1986. 28. Dean JA, Dean AG, Burton A, et al. Epi Info, version 3: Computer Programs for Epidemiology. Division of Surveillance and Epidemiologic Studies, Epidemiology Program Office. Atlanta: Centers for Disease Control, 1988. 29. SASISTAT Guide for Personal Computers, NC: SAS Institute, 1986.
version 6.0. Cary,
30. Hennekens CH, Burning JE, Mayrent SL, eds. Epidemiology in Medicine. Boston: Little, Brown and Co., 1987. 31. Weitzman M, Alpert JJ, Kierman LV, et al. High-risk youth and health: The case of excessive school absence. Pediatrics 1986;78:313-22. 32. Goolsby L, Principal at Alternative Education Program. Personal communication. January 30, 1991. 33. Office of State Budget and Management. Management Information Services, State Rankings 1986, first edition. Raleigh, NC: NC State Data Center, 1986. 34. Simmons J, Finway B, Yang A. The Adolescent and Young Adult Fact Book. Washington DC: Children’?, Defense Fund, 1991. 35. Klerman LV. School absence: A health perspective. Clin North Am 1988;35:1253-69.
Pediatr
MARCELLA
T. McCORD,
JANE M. FOY, M.D.,
M.D.,
M.P.H.,
JONATHAN
AND KATE FOTHERGILL,
School-based health clinics (SBCs) have been promoted as an innovative approach to providing adolescent health care. The present study examined the effect of a SBC on academic success. We studied the effect of clinic registration and use on students’ absence, suspension, withdrawal, and graduation or promotion rates in an alternative high school for students who were not able to succeed in traditional educational programs. On average, these 322 high-risk students attended school only 58% of the time; 24% were suspended; and only 26% graduated or were promoted. The 189 (59%) students who were registered to use the clinic and the 159 (49%) who actually used the clinic were as likely to be absent or to be suspended as non-registered students. However, students who used the clinic were significantly more likely to stay in school, and to graduate or be promoted than students who were not registered for the clinic. This relationship was strongest for black males; those who used the clinic were nearly three times [odds ratio = 0.35; 95% confidence interval (CI) = (0.16-0.78~1more likely to stay in school than those who did not use the clinic. In multiple linear regression models predicting school performance, only clinic use and percent of enrolled days absent were significantly associated with graduation/promotion, and these two variables predicted 23% of the variance in promotion status.
From the Departments of Maternaland Child Health (M.T.M.) and Health Behavior and Health Education (RF.), the School of Public Health, and The Cecil G. Sheps Center for He&h ServicesResearch (I.D.K.), The t,tniversity of North Carolina at Chapel Hill; the Guilford County Health Department, Child Health Division, Greensboro, North Carolina (J.M.F.) and the Division of AdolescentMedicine, Department of Pediatrics, Universitrr of Rochester School of Medicine, Rochester, New York (J.D.K.1. Address reprint requests to: ]onathan D. Klein, M.D., M.P.H., Department of Pediatrics, Division of Adolescent Medicine, Box 690, Dniversity of RochesterSchoolof Medicine, 601 Elmwood Avenue, Rockester, Neza York 14642. Manuscript accepted October 7‘1992.
0 Society for Adolescent Medicine, 1993 P&4ished by Elsevier Science Publishing Co., Inc., 655 Avenue of
D. KLEIN,
M.D.,
M.P.H.,
M.P.H.
KEY WORDS:
School-based
clinic
School performance Adolescent School dropout
The school dropout rate for adolescents in the United States in 1989 was 12.6% (l), and that for North Carolina in 1990-1991 was 5.6% (2). Adolescents drop out for a variety of reasons: boredom or academic failure, health problems (i.e., mental health crises, substance abuse, pregnancy), family concerns, financial considerations, or other problems (1). The transition to high school has been associated with decline in grade point average (3,4), and poorer school attendance (4); this may have long-term implications for adolescents’ adjustment (4), as low grades and poor attendance correlate highly with future school dropout (5-7). Adolescents who do not complete high school are at increased risk for multiple health, economic, and social problems including unemployment, welfare dependency, and homelessness (1). Schools are the only institution regularly attended by most young people ages 5 through 16 years; nearly 95% of all children and youth are in elementary or secondzry schools (8). An effective strategy for improving the health behavior of society is to focus on youth as they develop health habits and to provide adolescents with the knowledge, skills, and motivation for engaging in healthy behaviors (8). Thus, schools represent a public institution with great opportunity for improving the health of the school-age population (8). Comprehensive schoolbased health programs, including school-based health clinics (SBCs), have been successful at im91
the Americas, New York, NY 10010
92
McCORD ET AL.
proving adolescents’ access to health and related services (1,g); increasing utilization of health care; familiarizing students with the health-care system; and identifying and treating adolescents’ health problems (9). There are now more than 300 SBCs operating in middle schools and high schools across the United States(10). Many of the students served by these clinics are low-income minority youth, with limited access to other sources of care (11). SBCs are well used by students in the schools they serve, with utilization rates typically reaching over 75% of inschool adolescents (11,12). For about half of enrolled students, school-based clinics are their sole or primary source of health care (1l), suggesting that these services and sites are acceptable to adolescents (12). Most published articles on SBCs have either been descriptive (13-16) or have taken an advocacy position without empirical data (17-21). However, some programs have reported success in improving contraceptive use, decreasing pregnancy and substance use rates, preventing school dropout, and improving school attendance (22-25). Two studies report SBCs improved attendance and lowered dropout rates (15,25). One study of pregnant and parenting adolescents in four cities showed a trend toward higher graduation rates (25), and the American School Health Association has presented testimony showing that 38% of the students in one N.Y.C. school reported that their attendance had improved as a result of the clinic (22). Dryfoos (15), however, has noted that school principals and advocates often believe SBCs positively effect absenteeism and dropout, these claims have not been supported by published data. In 1987, the Gillespie Park Education Center (subsequently renamed the Alternative Education Program or AEP), an alternative public high school and middle school for Greensboro, North Carolina Public Schools’ students whose needs cannot be met in a traditional educational program, received a Robert Wood Johnson Foundation grants to develop a SBC (26). Our goal in this study was to identify scholastic outcomes associated with use of the GHC; specifitally, to evaluate whether or not clinic use had an effect on students’ absences, suspensions, dropping Out, and graduation or promotion. We hypothesized that students who used the school-based clinic would have fewer absences and suspensions, would be less likely to drop out of school, and would be more likely to be promoted or graduate.
JOURNAL 03 ADOLESCZNT HEALTH Vol. 14, No. 2
Methods Sample and Site The AEP serves grades 6 through 12; 78% of the students are rnzorities, and most are economically disadvantaged (27). The student population at the AEP is a high-risk population, having failed to succeed in traditional schools. More than half of the students live in single-parent households headed by females. Approximately 22% of the students are selfsupporting, and approximately 4% live independently from their parents. Over 30% of the student population are either pregnant or parenting a child. In 1986, 35% of the AEP students reported having used “hard” drugs (other than alcohol, cigarettes, and marijuana) in the past year. Chronic attendance problems, academic underachievem.ent, and serious behavior problems are frequent among the AEP students. In a 1986 survey, the AEP students’ health risks were significantly higher than those of other Guilford County students in 13 of 15 categories tested, including diet, dental hygiene, exercise, smoking, use of alcohol and drugs, unsafe driving practices, and stress (27). Before the Gillespie Student Health Project (GSHP) opened, more than 25% of the AEP families surveyed cited transportation and distance as obstacles to obtaining health care (27). Almost half also reported not being able to afford health care. The grant funds the GSHP, which consists of health education, health promotion, screening, and clinical services, all delivered via the Gillespie Health Center (GHC). The all-female staff of the clinic consists of a part-time physician, nurse practitioner, and clerk, and a full-time registered nurse, socia! worker, and clerk. The project is administered jointly by the Guilford County Department of Public Health, Moses Cone Hospital, and the Greensboro Public Schools. Data Collection We collected archival data from the AEP, GHC, and the Guilford County Health Department about school enrollment, attendance, suspension, drop out, and graduation/promotion for all students who attended the AEP during the 1990-1991 academic year. This was the clinic’s fourth year of operation. School and demographic data were abstracted from school attendance records. Clinic registration and utilization data, as well as additional demographic information, were obtained from the clinic. For an adolescent to be registered in the clinic, his or her
March 1993
SCHOOL-BASED CLINIC USE AND SCHOOL PERFORMANCE
parent must first have a face-to-face interview with clinic staff, either at school or during a home visit. Files were merged by name and confidentiality was maintained by removal of all student identifiers. The data were obtained through ongoing program evaIuation and confidential data-sharing agreements between the school and the Guilford County Health Department. Data were entered using Epi Info, Version 3 (28) directly into a portable computer from school, clinic and health department files.. Statistical Analysis
All analyses were conducted using the SAS statistical analysis system, version 6.0 (29) for personal computers. The statistical significance of observed differences between group mean values was done with the student’s t test. Analysis of variance was done using SAS general linear regression modeling to compare the mean values across multiple groups. This procedure was also used for multiple linear regression modeling to measure associations between the school-performance variables, clinic use, and demographic factors. For bivariate analyses, x2 tests were used to determine whether observed differences in proportions between study groups were statistically significant. Fisher’s exact test was used to compare differences between white student’s suspensions and promotions, as these cell sizes were quite small. Statistical significance for stratified data was determined using the Mantel-Haenszel test with one degree of freedom (30). In all cases, a p value of 0.05 was considered statistically significant.
Results During the academic year of August 1990-May 1991, 322 students were enrolled at the AEP. Mean time of enrollment was 102 days out of the 180-day academic year. Fifty-two percent of the AEP students were female, and 85% were black. The 189 students who were registered to use the GHC were similar to all enrolled students by gender, ethnic@, and age (Table 1). The population was also similar with respect to grade in school; however, because this is an alternative school, grade has little meaning. Because of this, age is used in the regression analyses. School Absence, Suspension, Promotion Status
and
Students were absent (including excused absences, unexcused absences, and suspensions) an average
93
Table 1. Summary of Enrolled Populationand Registered Population 1990-1991 Academic S&OO~ Year
Gender Female Male Ethnicity Black White Gender and ethnic@ Black female Black male White female White male Age (years) 12-13 14 15 16 17 18 19-20
School enrollees (n = 322) (B)
Cliic enrollees (n = 189’ (%) __
52 48
54 46
85 13
84 16
47 40 6 7
48 36 8 9
12 15 25 18 15 10 52
17 20 30 11 13 6 4
of 33.7 days. Excluding suspensions, AEP students were absent an average of 31.2 days. Nearly 20% of students were absent 50 or more days. When absences were considered as a proportion of the number of days enrolled in school, female students were absent a higher percent of time than males, and whites were absent a higher percent of time than blacks (Table 2). Overall, 78 (24%) students were suspended during the 1990-1991 school year. Among students who were suspended, the average number of days of suspension -was 2.5 days. Six students were suspended for between 21 and 40 days and one student was
suspended for 57 days. More black students were suspended than whites; and black students, on average, were suspended for a higher percentage of the time they were enrolled in school than were white students. Male students also were suspended a higher percentage of the time they were enrolled than were female students (Table 2). Black males were suspended an average of 6% of the time they were enrulled, compared with 1% for black females, white males, or white females. Eighty-four (26%) students were promoted or graduated during the 1990-1991 school year: 8 (3%) students graduated, 50 (16%) were promoted at the end of the year, and 26 (8%) were promoted midyear.
Nearly two-thirds (62%) of all students with-
94
JOURNAL OF ADOLESCENT HEALTH Vol. 14, No. 2
l&CORD ET AL.
Table 2. Average Percent of Time Absent and Suspended While Enrolled and Average Number of Clinic Visits
Time absent (%) Time suspended (96) Averape number of clinic visits
Female (n = 145)
Male (n = 123)
p value
Black (n = 226)
White (n = 37)
p value
42 1 6
35 5 4
0.005 0.0001 0.004
38 3 5
49 1 4
0.005 0.002 0.69
drew from school at least once during the academic year, Of these 43 (21.4%) m-entered school and 32
(74%) withdrew from school a second time. Nearly three-fourths of AEP students were retained in their grade in 1990-1991. On average, students who graduated or were promoted had fewer absences or suspensions as a percentage of their time enrolled in school than those students who were retained (Table 2). Males, females, blacks, and whites were all equally likely to graduate or be promoted.
Clinic Registration and Visits A total of 14!57visits were made to the clinic during the 1990-1991 school year. In addition to the 322 students enrolled at the AEP, 27 ex-students (dropouts, graduates, transfers to other schools) and 47 children of the AEP students (or ex-students) used the clinic. The students who were enrolled in the ABP for the 19!&1991 school year made 1060 visits.
Of these 322 students, 189 (59%) had registered to use the clinic, and 159 of these actually used the clinic one or more times. The average number of clinic visits for the 189 registered students was 4.65. Of the 189 registered students, 30 (16%) made no visits. About half (55.0%) of clinic re-:strants made one health-screening visit, and half (50%) of the registered students also made between c s?t and five treatment visits, 15% made 6-10 treat.nr;lt visits. Blacks were nearly twice as likely not to have registered as whites (44% versus 27%; p < O&Q, however, once registered, blacks and whites made the same average number of clinic visits (Table 2). Males and females were equally likely to be registered for the clinic. On average, however, females made more clinic visits than males (Table 2). Clinic Use and School Performance Overall, students who registered for the clinic were enrolled in school an average of 112 days. Those who were not registered to use the clinic were enrolled an average of 92 days (p < 0.002). The percent of time students were absent while they were enrolled did not differ for students registered and not
Retained Promoted (n = 186) (n = 82) 44 4 3
26 2 8
.-
p value o.OOBO1 0.025 o.OOB1
registered for the clinic (38% versus 41%; p = 0.19). However, students who actually used the clinic were absent a lower percent of the time they were enrolled in school (36%), compared to students who registered but did not use the clinic (51%) and students who did not register (41%) (p C 0.02). More students who were registered for the clinic were suspended than students who were not registered (30% versus 16%) (x2 = 8.781, p < 0.003).
Controlling for race and gender (Table 3), we found that black students who had registered were twice as likely to be suspended than those not registered, and black males were nearly three times as likely to be suspended if registered for the clinic. Black male and black female students who registered and used the clinic were even more likely to be suspended than those who registered but never made a visit. However, there is no difference between the actual percent of time spent suspended while enrolled in school for registered students and students who registered but did not make a visit. Students who were registered for the clinic were more likely to stay in school than students who were
Table 3. Adjusted Odds Ratios Comparing Registered
and Non-RegisteredStudent Odds of Withdrawal from School, BeingSuspended, and Being Promoted (By Race and Gender) Withdrawal
All0 Blacks Whites Females Males Black females
Black males White females White males
0.5** 0 4** 1:9 0.6 0.4* 0.5* 0 3** 1:2 2.2
Suspension 2.3” 2 3** 0:o 2.7, 2.4* 2.4 2.8’+ 0.0 0.0
Promotion 1.8* 2.1” 0.7 2.0 1.5 2.3’ 1.7 0.5 0.8
“Referencecategory including all students at the school (i.e., for all students, those who were registered were half as likely to withdraw from school as those not registered). p value < 0.05. l*p value < 0.01. N.B.: The associations between the study variables vanes slightlv by the different strata of the control variables, however, this variance is not supported by a significant Breslow-Day test for any of the variables.
SCHOOL-BASED CLINIC USE AND SCHOOL PERFORMANCE
March 1993
Table 4. Percentages of Students Who Stayed in School and Were Promoted Studentswho stayed in school (%) Female (168) Male (154) Black (274) White (41) Total (322) Students
promoted
Female (168) Male (154) Black (274) White (41) Total (322)
Registered Not registered for clinic
for clinic
p value
40 49 48 23 44
28 29 28 36 29
0.10 0.02 0.001 0.044 0.005
36 24 33 20 31
22 18 19 27 20
0.05 0.30 0.00 0.68 0.02
(a)
not registered (44% versus 29%) (x2 = 7.835, p < 0.005). Almost half (48%) of black students registered for the clinic stayed in school compared to 28% of non-registered black students (Table 4). Registered male students were also twice as likely to stay in school than non-registered students. Controlling for both race and gender (Table 3), black males were nearly three times as likely to stay in school if they had registered for the clinic. Black female students who were registered were also more likely to stay in school than those not registered. Among whites, registered students were no more likely to stay in school than those not registered. When we examined the likelihood of staying in school by whether students actually used the clinic services, we found that students who registered and used the clinic were more likely to stay in school than those who registered but never made a visit. Almost half (48%) of those students who used the clinic (compared to only 23% of registered students
95
who did not make clinic visits) stayed in school. This relationship was striking among black males; 60% of those who registered and used the clinic stayed in school compared to only 18% of those who were registered but did not use the clinic (Figure 1). Students who were registered were also more likely to graduate or be promoted than students who were not registered (31% versus 20%) (x2 = 4.86, p < 0.03). Students who were registered were nearly twice as likely to graduate or be promoted [odds ratio (OR) = 1.81; 95% confidence interval (CI) = (1.067-3.054)]. Students who graduated or were promoted averaged eight clinic visits compared to three visits made by students who were retained in their grade (p < 0.0001) (Table 2). Among blacks, 33% of students who registered for the clinic graduated or were promoted compared to 19% of non-registered students (Table 4). Among whites, registered students were no more likely to graduate or be promoted than those who were not registered. Students who registered and actually used the clinic services were more likely to graduate or be promoted than those who registered but never made a visit. Nearly two-thirds of all black males who graduated or were promoted had used the clinic (Figure 2).
uItivariate
Linear Regression
We used multiple linear regression models (SAS GLM procedure) to assess the strength of the association among clinic visits, and absences and promotion or graduation status. In these analyses, the total number of clinic visits was the independent variable.
60
Percentwho my6d in school
40
who W.R promoted
Black Female
Figure 1. Percent of clinic registrants, users, and non-users who stayed in school by race/gender groups. Overall p = 0.077 for black females, p = 0.001 for black males, p = 0.857for white females, and p = 0.592 @ white males.
Figure 2. Percentage of clinic registrants, users, and non-users who graduated or were promoted by race/gender groups. Qoerall p = 0.034 for blackfemales and p = 0.034 for black nudes. Too fewwhite students were graduated to de$ct in this figure.
96
JOURNAL OF ADOLESCENTHEALTH Vol. 14, No. 2
McCORD ET AL.
The independent variables did not determine very much of the variance for percent of time absent (tz = 0.083) or percent of time suspended (? = 0.099). However, the model accounted for a substantial portion of the variance in student’s promotion status (J = 0.233) (Table 5). In a stepwise model, the only two significant predictors of promotion status were percent of time absent (partial 3 = 0.180; p = 0.0001) and total number of clinic visits (partial ? = 0.047; p = 0.0001). Chances of promotion were increased by fewer absences and a higher number of clinic visits.
Discussion We expected that students involved with the schoolbased clinic at the AEP would have fewer absences or suspensions, would be more likely to stay in school, and would be more likely to graduate or be promoted to the next grade. Our results partially supported our hypotheses. For example, we found no difference in the days absent or suspended while enrolled between students registered and not registered for the clinic. Stud.ents who used the clinic were only slightly less likely to be absent than nonusers. However, registered students and students who actually used the clinic were twice as likely to stay in school and nearly twice as likely to graduate or be promoted than non-registered students. Additionally, the greater students’ exposure to the clinic (i.e., actual visits to the clinic), the stronger all of the relationships between clinic use and graduation or promotion became. The relationship between clinic use and staying in school was particularly striking among the black males at Gillespie. Males, especially those in minority groups, are traditionally the most difficult group to reach with adolescent health services (1,ll). Black males also drop out of school at higher Table5. Regression Models and B Estimates for Student’s Absences, Suspensions, and Promotion Time absent Time suspended WI (%I Total 9 (F;p value) Age Gender Ehicity Clinic visits Percent absences Percent suspensions
(s.8llzL1, -0.704 -5.262 0.230 -1.265
(7.1ozLo1) -0.443 4.221 -3.280 NS
Promotion or graduation (r2.9gzIoo*) - 0.010 -0.068 - 0.013 0.020
-
-
-0.008
-
-
NS
than either black females or whites (1). In this population, black males who used the clinic were three times as likely to stay in school as black male clinic non-users, and two-thirds of black males who graduated or were promoted were clinic users. The entire student body at the AEP is at high risk for school failure. Thus, unfortunately, we were unable to study the way in which students who registered differed from those not registered. Part of the difference may lie in selection bias in the adolescents themselves. While this group may be more self-motivated and desire better access to health services, this might also have been expected to affect school attendance, which is similar for clinic registrants and non-registrants. Another possibility is that these students may have more health problems and therefore may need clinic services more than others. If health status among clinic users is worse than that of non-registrants, this might help explain absenteeism being similar and not lower among the clinic population. A third factor may be attributable to the adolescents’ parents. For an adolescent to be registered in the clinic, his or her parents must have had a faceto-face interview with clinic staff, either at school or at a home visit. Thus, clinic registration may reflect the parents’ involvement in their child’s education. A final factor may be attributable to the school staff. School authorities noted that the school and clinic staff may, in fact, try hardest to reach students at highest risk. For both males and females, the students at highest risk in an already high-risk population may be targeted to receive the greatest number of services. Students who used the clinic averaged a slightly lower percent of time absent than non-users and ‘or the clinic. Ninetythose students not register four percent of the AEP shtdents were absent for one day or more, and 70% were absent more than 14 days. In other North Carolina schools, only 9% of students are absent for more than 14 days (2). In Guilford County, this figure is 8% (2). In other innercity school, estimates of daily school absence rates are between 20% and 40% (7,31). The students at the AEP were absent slightly over two-fifths of the total amount of time they were enrolled. Another limitation of this study is that we were unable to differentiate between absenteeism for “health” reasons or for other reasons. This is particularly true for pregnant adolescents who are “homebound,” who are counted absent at the AEP even though they are receiving home curricula. This may account for up to 200 days absent per term (32). Additionally, rates
March 1993
SCHOOL-BASED CLINIC USE AND SCHOOL PERFORMANCE
clinics cannot prevent many of the illnesses that might keep students home, and clinic staff may in fact facilitate absences by correctly diagnosing illnesses that should be treated by bed rest (11). These two factors might help explain the lack of difference in percentage of time absent between clinic users and non-users, at least among females. We are encouraged by the fact that the chance of staying in school is increased by students’ contact with the clinic. Overall, 62% of the population of students at the AEP enrolled for the 1990-2991 academic year withdrew from school. In North Carolina, 45% of persons 25 years or older in 1980 reported not having completed high school (33). The dropout rate for North Carolina for 1990-1991 was 5.6%, and for Guilford County it was 3% (2). In 1989, only 52% (56% of whites and 32% of blacks) of high school dropouts were employed, whereas 76% (79% of whites and 60% of blacks) of high school graduates were employed (34). Although the chances of adolescents at the AEP staying in school is low, students having greater contact with the clinic were more likely to stay in school. The reasons that these adolescents stay in school cannot be understood from this study. The AEP and the GHC aim’ to provide a supportive environment that will encourage teenagers to stay and finish high school. While use of the clinic is unlikely to be the actual cause of better school performance, a strong relationship exists between number of clinic visits and school promotion. A possible explanation is that the trust and support provided by SBC personnel helps students identify with and function better in school. We believe that SBCs provide positive role models and social support for the adolescents they serve. By having accessible, affordable, and confidential services at the school, students are provided not only with a convenient source of health care. They also may in fact gain a source of increased social support. Students may benefit from both the services rendered at the clinic and the perception that support services are available. Because we were limited to secondary data, we did not attempt to measure actual or perceived social support. However, this conceptual model is useful in helping to explain our findings. In general, excessive absenteeism pre diets those adolescents who are at highest risk for school drop-out and failure (35). In our population,
there is a school-wide problem with excessive absence, but some of these adolescents stay in school, and these are the students who have used the SBC. These students are not only staying in school, but
97
they are also succeeding academically, successfully moving ahead to higher grade levels, and fulfilling requirements for graduation. Understanding these phenomena will require prospective studies of students at schools with and without school-based clinics. This study was supported in part by the School-Based Health Program of The Robert Wood Johnson Foundation and by grant R48CCR402177 from the Centers for Disease Control. The authors thank Julia G. Lear and Elizabeth McAnamey for commenting on earlier drafts of this manuscript, Carol Porter for computer programming advice, and the staff of the Greensboro Public schools’ Alternative Education Program, the Gillespie Health Center and the Guilford County Department of Public Health for their assistance in data collection.
References 1. Office of Technology Assessment. Adolescent Health Volume 1. Summary and Policy Options. United States Congress, 1991. 2. The North Carolina Public Schools 1991 Statistical Profile. Annual publication. Raleigh, NC: N.C. Department of Public Instruction, 1991. 3. Blyth DD, Simmons RG, Carlton-Ford S. The adjustment of early adolescents to school transitions. Br Educ Res J 1983;8:149-64. 4. Felner RD, Faber SS, Primvera J. Transitions and stressful life events: A model for primary prevention. In: Felner RD, Jason LA, Moritsugu JN, et al, eds. Preventive Psychology: Theory, Research and Practice in Community Intervention. New York: Pergamon, 1983. 5. U.S. Department of Health, Education, and Welfare. Dropout Prevention. Washington, DC: Education Resources Information Center, 1975. 6. Weitzman M, Klerman LV, Alpert JJ, et al. Factors associated with excessive school absence. Pediatrician 1986;13:74-80. 7. Klerman LV, Weitzman M, Alpert JJ, et al. Why adolescents do not attend school: The views of students and parents. J Adolesc Health Care, 1987;8:425-30. 8. Kirby D. Comprehensive school health and the larger community: Issues and a possible scenario. J Sch Health 1990;60:170-7. 9. Council of Scientific Affairs. Providing medical services through school-based health programs. J Sch Health 1990;60:87-91. 10. Waszak C, Neidell S. School-Based and School-Linked Clinics: Update 1991. Washington, DC: Center for Population Options, 1991. 11. Kirby D, Waszak CW, Ziegler J. An Assessment of Six SchoolBased Clinics: Services, Impact and Potential. Washington, DC Center for Population Options, 1991. 12. Klein JD, Slap GB, Elster A, et al. Access to health care for adolescents: A position paper for the Society of Adolescent Medicine. J Adolesc Health Care 1992;13:162-70. 13. Blum RW, Pfaffinger K, Donald WB. A school-based comprehensive health clinic for adolescents. J Sch Health 1982;52:486-90. 14. Edwards LE. Steinman, ME, Hakanson, EY. An experimental comprehensive high school clinic. Am J Public Health 1977; 673765-6.
98
JOURNAL OF ADOLESCENTHEALTH Vol. 14, No. 2
McCORD ET AL.
15. hyfoos J. School-based health clinics: A new approach to y;nz;tinn5adolescent pregnancy? Fam Plann Perspect ;: . 16. Kirby D. Comprehensive school-based health clinics: A growing mcvement to improve adolescent health and reduce teenage pregnancy. J Sch Health 1986;56:289-91. 17. Cherskov M. SchooI clinics help kids and hospitals, too. Hospitals 1986;60:95. 18. Davis JM, Devaney, JM. Comprehensive centers. West J Med 1986;144:625-6.
school-based
teen
19. Gonzales C, Mu!liganD, Kaufman A. Adolescent health care: Improving access by school-based services. J Fam Pratt 1985;21:26%70. 20. Keenen T. School-based adolescent Pediatr Nurs 1966;12:365-9.
health care programs.
21. Lisoskie S. Health clinics in high schools. J Am Med Worn
Aasoc 19&5;41:106-126. 22. American School Health Association. Testimony on the Year 2ooo Objectives for the Nation: Recommendations for the School Health Program. Dayton, OH: April, 1988. 23. Dryfoos JG. School-based health clinics: Three years experience. Fam Plann Perspect 1988;20:193-200. 24. Joiner T. The role and success of school-based Ford Hasp Med J 1986;36:230-1.
clinics. Henry
25. Zabin LS, Hirsch MB, Smith EA, et al. Evaluation of a pregnancy-prevention program for urban teenagers. Fam Plann Perspect 1986;18:119-26.
26. Robert Wood Johnson Foundation. The School-Based Adolescent Health Care Program. Washington, DC: April, 1986. 27. Guilford County Health Department. Needs Assessment for Adolescent Health in Guilford County, 1985-86. Greensboro, NC. 1986. 28. Dean JA, Dean AG, Burton A, et al. Epi Info, version 3: Computer Programs for Epidemiology. Division of Surveillance and Epidemiologic Studies, Epidemiology Program Office. Atlanta: Centers for Disease Control, 1988. 29. SASISTAT Guide for Personal Computers, NC: SAS Institute, 1986.
version 6.0. Cary,
30. Hennekens CH, Burning JE, Mayrent SL, eds. Epidemiology in Medicine. Boston: Little, Brown and Co., 1987. 31. Weitzman M, Alpert JJ, Kierman LV, et al. High-risk youth and health: The case of excessive school absence. Pediatrics 1986;78:313-22. 32. Goolsby L, Principal at Alternative Education Program. Personal communication. January 30, 1991. 33. Office of State Budget and Management. Management Information Services, State Rankings 1986, first edition. Raleigh, NC: NC State Data Center, 1986. 34. Simmons J, Finway B, Yang A. The Adolescent and Young Adult Fact Book. Washington DC: Children’?, Defense Fund, 1991. 35. Klerman LV. School absence: A health perspective. Clin North Am 1988;35:1253-69.
Pediatr