Perceived Knowledge and Training Needs in Adolescent Pregnancy Prevention: Results From a Multidisciplinary Survey | Adolescent Medicine | JAMA Pediatrics | JAMA Network
[Skip to Navigation]
Sign In
July 2000

Perceived Knowledge and Training Needs in Adolescent Pregnancy Prevention: Results From a Multidisciplinary Survey

Author Affiliations

From the Maternal and Child Health Program, Division of Epidemiology, School of Public Health (Dr Hellerstedt, Ms Smith), the National Teen Pregnancy Prevention Research Center, Schools of Medicine, Nursing, and Public Health, (Drs Hellerstedt, Shew, and Resnick and Ms Smith), and the Division of General Pediatrics and Adolescent Health, School of Medicine (Drs Shew and Resnick), University of Minnesota, Minneapolis. Dr Shew is now with the Department of Pediatrics, Indiana University School of Medicine, Indianapolis.

Arch Pediatr Adolesc Med. 2000;154(7):679-684. doi:10.1001/archpedi.154.7.679

Objectives  To examine health care professionals' knowledge and interest in training in adolescent pregnancy prevention and whether an association exists between perceived knowledge and interest in training.

Design  A cross-sectional mailed survey.

Participants  Random, stratified sample design that identified 800 psychologists, 800 social workers, 1000 nurses, and 400 pediatricians from national professional membership lists. Response rate to the mailed survey was 51%. After removing respondents who did not currently work with adolescents, 1242 surveys (41%) were available for analyses.

Main Outcome Measures  Descriptive analyses were conducted on self-report data concerning perceived knowledge and interest in training about adolescent pregnancy prevention separately for each of the 4 disciplines. Within disciplines, perceived knowledge and interest in training were correlated for each of 3 content areas (ie, sex education and contraceptive counseling, adolescent pregnancy, and counseling after a negative pregnancy test) and for a summary measure of the content areas.

Results  Less than half of the nursing, pediatrics, psychology, and social work professionals reported high perceived knowledge in the 3 content areas. Psychologists and social workers reported the lowest perceived knowledge. However, with the exception of psychologists, more than two thirds of the other respondents reported moderate or high interest in training in the 3 content areas. Interest in training was not strongly correlated with perceived knowledge within any discipline.

Conclusions  The need to integrate psychosocial components into adolescent health care is a core assumption in the field, yet these data indicate that psychologists and social workers perceive low levels of knowledge and interest in training. These disciplines may benefit from more targeted professional training about their role in preventing adolescent pregnancy.

RECENT DATA show encouraging trends in adolescent pregnancy: birth rates have decreased in all 50 states as well as the District of Columbia1 and abortion rates for 15- to 19-year-olds have also declined, from 42.9 per 1000 in 1981 to 32.2 per 1000 in 1994.2 These reductions have been attributed to several factors, including a decrease in the proportion of adolescents who are sexually active, an increased likelihood that adolescents will use contraception, especially condoms, at first intercourse, and the use of injectable and implantable contraceptives, especially among black youth, who have historically had the highest adolescent pregnancy and birth rates.1 Declines in adolescent pregnancy rates should fuel optimism but not complacency, because the number of pregnancies in females younger than 20 years in the United States continues to approach 1 million per year; more than three quarters of these pregnancies are assumed to be unintended.2 Also, in the near future, we may see an increase in adolescent pregnancy and childbearing. It is estimated that between 1995 and 2005, the number of 15- to 17-year-olds in the United States will increase by 15%; the greatest increases will be in the nonwhite population, who have 2 to 3 times higher fertility rates than whites.3 If the current adolescent fertility rate remains the same, it is estimated that there will be a 26% increase in the number of adolescent pregnancies.4 Our concerns about adolescent pregnancy and parenting, then, are likely to remain high on the health agenda for the nation.

The attribution of recent declines in adolescent pregnancy to lower numbers of sexually active adolescents and to better contraception among sexually active adolescents reflects the importance of effective sex education and contraceptive counseling. Kirby5 underscores this in his finding that the provision of contraceptives alone by school clinics without a strong educational component may not increase contraceptive use or decrease pregnancy rates in adolescents. In addition to effective sex education, targeting high-risk adolescents for intervention may also prevent pregnancy. For example, Zabin and colleagues6,7 identified adolescents who receive negative pregnancy test results from health care providers to be a particularly accessible and high-risk group because they are sexually active, likely to be exposed to ongoing risk for pregnancy, and available for intervention. In their small longitudinal study of adolescents who were 17 years or younger when they received a negative pregnancy test, Zabin et al7 found 58% were pregnant within 18 months of the pregnancy test. To test the generalizability of their findings, they conducted a survey of adolescents who presented for pregnancy tests in 52 US clinics. The results of this survey of almost 3000 females who were 17 years or younger showed that, among those who had positive pregnancy tests, 25% had had a previous negative pregnancy test at a clinic site.6

It is generally agreed that adolescent pregnancy, and particularly childbearing, are medical concerns with compelling social antecedents and consequences. It is the social costs of adolescent childbearing, in fact, rather than the medical or health consequences, that have stimulated the work of many clinicians, researchers, policymakers, and educators.4,5,8-12 The individual, social, and economic conditions of adolescents who become pregnant, and particularly those who become mothers, are strikingly distinct from those of adolescents who do not become pregnant.4,5,11 Given the number and variety of antecedents of adolescent pregnancy, it is clear that a multidisciplinary approach is necessary for effective intervention. Antecedents of adolescent pregnancy exist at community, family, and individual levels and include biological, psychosocial, economic, and political factors.4,5,8,10,11 While physicians and nurses have traditionally played an important part in adolescent pregnancy prevention efforts, professionals in psychology and social work also have a role in prevention efforts. Adolescent sexual risk-taking may co-occur with other risk factors, such as family problems, school failure, and conduct disorder4,5,8,11-14; thus, psychologists and social workers may be the first providers to have contact—or may have the most sustained contact—with adolescents at highest risk for pregnancy. Adolescent health programs, and especially pregnancy prevention programs, have evolved to address issues that fall more into the domains of psychologists and social workers, including youth development, resilience, family and adult connectedness, and job- and school-readiness.5,12,13 Correspondingly, the sites for adolescent health promotion programs have moved beyond medical and health care settings to community-based sites, including schools, churches, and youth centers.11-13

Several studies suggest that some professionals may not be adequately prepared to address the sexual or reproductive health needs of youth.15-19 Respondents to a national survey of 351 primary care physicians reported high levels of perceived deficits in competence related to adolescent health concerns, especially high-risk behaviors.16 In another national survey of 3066 physicians, nurses, psychologists, social workers, and nutritionists, respondents reported perceived deficits in training and competence in various areas of adolescent health care. For example, more than half of the psychologists and social workers reported deficits related to sexual concerns and sexual orientation.18 Surveys do not consistently show that health providers desire training in adolescent family planning or sexual concerns.16 For example, a survey about the perceived continuing education needs of 331 Indiana physicians in a variety of practices showed that subjects had minimal interest in providing family planning services, or receiving training about family planning, for their adolescent patients.19 The findings from these surveys parallel research findings that physicians fail to consistently screen adolescents for high-risk behaviors. In a sample of 343 California physicians, Ellen et al20 found that while physicians often screened adolescents for substance use and sexual activity, older adolescents were more likely than younger adolescents to be screened and there were differences in screening by specialty. In addition, a survey of more than 1200 California physicians showed that only 40% of the physicians screened all of their adolescent patients for sexual activity; even lower percentages screened for specific sexual behavior such as number of partners, sexual orientation, and frequency of casual sex.21 Of interest was that higher screening was associated with more recent date of medical school graduation.

The objective of this study was to assess perceived knowledge and interest in training about adolescent pregnancy prevention among professionals in nursing, pediatrics, psychology, and social work, and to examine the association between present knowledge and interest in training within these disciplines. This needs assessment was conducted to identify future training directions for enhancing the capacity of adolescent health care providers in pregnancy prevention.

Participants and methods


The National Needs Assessment of Adolescent Health Care Providers was a mail survey that was administered in 1997 at the University of Minnesota, Minneapolis. The survey was derived, in part, from a national needs assessment of adolescent health professionals conducted by the University of Minnesota in 1986.18 A multidisciplinary research team finalized the discipline-specific surveys for the 1997 survey to assess the perceived knowledge of adolescent health professionals in 4 disciplines: nursing, pediatrics, psychology, and social work. Surveys were pilot-tested with 10 members of each discipline and their feedback was incorporated into the final drafts. The sample frame for the surveys was constructed with membership lists from the following professional organizations: (1) the National Association of Pediatric Nurse Associates and Practitioners, the National Association of School Nurses, and nurse members of the American Public Health Association; (2) the American Academy of Pediatrics; (3) the American Psychological Association; and (4) the National Association of Social Workers. Stratified random sampling was used to identify 3000 members of these organizations. Based on the size of the membership lists, 800 surveys were sent to practitioners in psychology, 800 surveys were sent to social work professionals, 1000 surveys were sent to nurses, and 400 surveys were sent to pediatricians. A reminder postcard was sent 2 weeks after the first survey was mailed. A second copy of the survey was sent 3 weeks later to those who did not respond to the first mailing and the reminder postcard. The overall response rate was 51%. After removing those who retired or whose current practice did not include adolescents, 1272 surveys (42%) were available for the current analysis. Table 1 shows the response rates and the final sample available for analysis for each discipline.

Table 1. 
Response Rate and Final Sample Size of the National Needs Assessment of Adolescent Health Care Providers
Response Rate and Final Sample Size of the National Needs Assessment of Adolescent Health Care Providers


Data were collected on several demographic and practice variables, including highest degree completed, years since last degree, years worked with adolescents, percentage of clinical population composed of adolescents, and practice setting. Respondents were also asked to assess their present knowledge and future interest in training for numerous conditions or concerns affecting adolescent health (eg, pregnancy, obesity, substance abuse). For each condition or concern, respondents were asked to indicate their level of knowledge (ie, low, moderate, or high knowledge level) and their interest in training (ie, low, moderate, or high interest). While the survey examined several adolescent health concerns, the present study is restricted to descriptive analyses of 3 intervention areas related to adolescent pregnancy: sex education/contraceptive counseling, adolescent pregnancy, and counseling adolescents who test negative for pregnancy.

Perceived knowledge data were collapsed into 2 categories: low/moderate vs high knowledge, with the rationale that respondents indicating less than high knowledge were potentially more likely to benefit from additional training. Data about interest in training were collapsed into low vs moderate/high interest, based on the rationale that those expressing either moderate or high interest in training might be motivated to obtain training, whereas those who indicated low interest probably would not. Respondents were also given a list of several continuing education methods and asked to indicate how useful they considered each method.


Descriptive analyses were conducted separately for each of the 4 disciplines. Perceived knowledge and interest in training were examined using χ2 analysis for each content area (ie, sex education/contraceptive counseling, adolescent pregnancy, and counseling adolescents who test negative for pregnancy) and in relation to the demographic and practice variables. In addition to examining the content areas individually, items for the 3 content areas were also combined to form 2 summary scales: a perceived knowledge scale and an interest in training scale (Cronbach α for each scale, ≥.73). The correlations between the perceived knowledge and interest in training scales were examined using Pearson correlation analysis.


Respondents varied in the number of years since they had completed their highest degree, the number of years they had worked with adolescents, and the percentage of their patient or client population that was adolescent (Table 2). Many respondents reported practicing in multiple settings: private practice was reported most often by psychologists (81%) and pediatricians (43%); nurses most commonly reported practicing in school-based settings (42%); and child and family services was reported most often by social workers (25%).

Table 2. 
Education and Practice Characteristics of Respondents by Discipline
Education and Practice Characteristics of Respondents by Discipline

Table 3 summarizes data on perceived knowledge and interest in training for each of the 3 content areas across the 4 disciplines. With the exception of pediatrics, less than half of the respondents from each discipline reported high knowledge in areas related to adolescent pregnancy prevention. The percentage of respondents reporting high knowledge varied: respondents from the discipline of psychology reported the lowest percentage, followed by social work. In contrast to the perceptions about knowledge, more than half of the respondents from nursing, pediatrics, and social work reported moderate to high interest in training in the content areas, with more than 70% of the respondents from each discipline reporting moderate to high interest in at least 1 area of pregnancy prevention. Counseling for a negative pregnancy test was reported as the content area of lowest knowledge and lowest interest in training by members of all disciplines except pediatrics.

Table 3. 
Percentage Reporting High Knowledge and Moderate or High Interest in Training by Discipline and Content Area
Percentage Reporting High Knowledge and Moderate or High Interest in Training by Discipline and Content Area

In correlational analysis of the summary content scores within discipline, perceived knowledge was not associated with interest in training for nursing or pediatrics. However, there were modest positive correlations between perceived knowledge and interest in training for social work (r=0.18, P=.004) and for psychology (r=0.29, P=.001). There were no differences in perceived knowledge and interest by demographic or practice variables.

Table 4 describes the perceived utility of a variety of continuing education methods. Responses were consistent across disciplines, with small conferences and workshops rated as the most useful methods of training and Internet and teleconferencing/interactive television methods rated as the least useful.

Table 4. 
Percentage Who Consider Continuing Education Method to Be Very Useful By Discipline
Percentage Who Consider Continuing Education Method to Be Very Useful By Discipline


This appears to be the first survey of a national sample to specifically examine perceived knowledge and interest in training in adolescent pregnancy prevention across disciplines, although other needs assessments concerning adolescent health care have been reported.15,16,18,19 The data reveal that there is interest in training opportunities in adolescent pregnancy prevention. Lower levels of perceived knowledge by the disciplines of psychology and social work as well as less perceived need for training by respondents from psychology suggest that these disciplines may benefit from more targeted professional training about their roles in preventing adolescent pregnancy. Psychologists reported the lowest perceived knowledge and the lowest interest in training for all 3 content areas; psychologists also exhibited the strongest positive association between perceived knowledge and interest in training. These findings suggest members of this discipline not only lack knowledge, but may perceive this knowledge as outside of their domain. The implications of these data may be that training needs to be preceded by education about the potential importance of behavioral and social scientists in adolescent pregnancy prevention.

Interest in training was not correlated with perceived knowledge for nurses or pediatricians. This finding suggests that, to be most effective, training should be tailored (from rudimentary to advanced) to match varying degrees of experience in these areas. And, as with the other disciplines, there may be professionals in nursing and pediatrics who not only report low knowledge, but perceive little need for this knowledge, as indicated by their low interest in training (ie, desire for training is not driven by present knowledge, but rather by perceived relevance). If the self-report data were not valid, the weak association between knowledge and interest in training could reflect methodologic limitations rather than a true absence of association, because invalid responses result in misclassification that could attenuate the observed associations.

There was lower perceived knowledge and interest in training about counseling adolescents who test negative for pregnancy across all disciplines except pediatrics. This finding suggests a need to inform professionals about the possibly high risk for pregnancy among adolescents who receive negative pregnancy tests and thus the potential opportunity for intervention with them. Given that Zabin et al6 estimated that 1 in 4 pregnancies may be prevented or delayed by counseling adolescents with negative pregnancy tests, this is arguably an important area to include in future training for professionals who work with adolescents. Because of the environmental, social, and psychological correlates of adolescent sexual activity and pregnancy,4,10,11 it is appropriate to include the specialized skills of psychologists and social workers, as well as physicians and nurses, in interventions with adolescents who receive negative pregnancy tests.

Similarities across disciplines regarding the most useful methods of continuing education are noteworthy. Overall, respondents reported that opportunities to interact with other professionals in relatively small numbers was more useful than activities done either individually or in larger groups. As technology evolves, many academic institutions are investigating electronic education methods for cost effectiveness and utility. This needs assessment suggests that these methods might not be as well-received as more personalized, hands-on training opportunities. This finding prompts a further question: what are the incentives for busy professionals when they seek continuing education? The role of convenience in obtaining continuing education is understudied; it is unknown how effective strategies to promote the professional salience of any topic area could be. The finding that more hands-on, face-to-face education was desirable could reflect the learning preferences of the respondent, but it could also reflect the perception that the most effective continuing education programs may be those that are physically separated from the distractions of the home and/or office because they provide an opportunity to focus on new materials.

The limitations to the current study inform recommendations for future needs assessments. First, the response rate of 51% potentially limits the generalizability of the results. Although there was wide variation in respondents' practice settings and other characteristics, it is difficult to ascertain whether respondents were representative of the larger population of professionals who serve adolescent clients. Low response rates have also been a problem with the few previous surveys of professionals regarding training needs in child and adolescent health. Of 4 studies with similar sampling frames to the current study and similar mailed survey protocols, response rates ranged from 72% among professionals in medicine, nursing, nutrition, psychology, and social work,18 56% among nurses,15 and 47% and 39% among physicians.16,19 Unfortunately, like the current study, none of these studies could characterize nonresponders, thus the degree and the direction of the potential selection bias is unknown. It is possible that those who responded to the current needs assessment were more likely than those who did not respond to perceive gaps in their knowledge. Conversely, those who felt they had adequate knowledge may have felt more comfortable responding. Therefore, perceived knowledge and interest in training may be overestimated or underestimated in this study. The results of this and similar studies suggest that querying by mail may not be the best method, although it is inexpensive and quick. Other methods, such as random telephone calling or in-person survey administration, could be employed to determine whether their additional costs are outweighed by more precise and representative data.

A second limitation of the study is that the validity of self-reported knowledge by professionals may be low,19,22,23 and the level of perceived knowledge reported in this study may not precisely reflect actual knowledge. However, assessing interest in training was also a goal in this study and perceived knowledge may be an important motivator for desire for training, although the correlations in the present study were not high.

A third limitation of the study is that, while the data indicated a perceived need for training, the data do not provide direction about how to meet those needs. Future assessments should address more specific, relevant content areas, incentives to obtain continuing education, and opportunities to facilitate referral among providers. While sex education, contraceptive counseling, and counseling adolescents who test negative for pregnancy are important components of pregnancy prevention efforts, it would also be useful to ascertain providers' knowledge of related issues, such as the antecedents of adolescent pregnancy, adolescent confidentiality and consent laws, and cultural issues associated with sex education and contraceptive use. Assessment is also needed to gauge knowledge about basic, related topics, such as behavioral, psychological, and cognitive adolescent development and reproductive physiology. More in-depth examination of these topics through instruments focusing solely on adolescent pregnancy prevention is recommended. It may also be worthwhile to query other professionals who counsel youth (eg, school counselors, clergy, teachers) rather than limit inquiry to health care providers. In addition, because adolescent pregnancy is a salient social and medical concern, perhaps no single discipline can exclusively provide comprehensive therapeutic intervention.18 It may be appropriate to assess interest in, and need for, training in risk assessment, brief interventions, and referral capacity.

While these data have yielded some information about how to strengthen the roles of nurses, physicians, psychologists, and social workers in adolescent pregnancy prevention efforts, further assessment is needed to identify relevant educational content and understand how to best target these disciplines—and possibly other professionals—for future training.

Accepted for publication February 10, 2000.

This study was supported in part by grants to the Maternal and Child Health Training Program (MCJ000111), the Graduate Studies in Adolescent Nursing Program (MCJ279185), the Adolescent Health Training Program (MCJ00985), and the Public Health Nutrition Training Program (MCJ273A03-03-0), University of Minnesota, Minneapolis, from the Maternal and Child Health Bureau, Washington, DC; and to the National Teen Pregnancy Prevention Research Center, University of Minnesota, Minneapolis (488-CCU513331), from the Centers for Disease Control and Prevention, Atlanta, Ga.

Presented as a poster at the annual meeting of the Society for Adolescent Medicine, Los Angeles, Calif, March 17-21, 1999.

We thank Marjorie Ireland, MS, for conducting the analyses for this article.

Corresponding author: Wendy L. Hellerstedt, MPH, PhD, University of Minnesota, School of Public Health, Division of Epidemiology, 1300 S 2nd St, Suite 300, Minneapolis, MN 55454-1015 (e-mail:

Ventura  SJCurtin  SCMathews  TJ Teenage Births in the United States: National and State Trends, 1990-96.  Hyattsville, Md National Center for Health Statistics1998;DHHS publication (PHS) 98-1019.
Henshaw  SK Unintended pregnancy in the United States.  Fam Plann Perspect. 1998;3024- 2946Google ScholarCrossref
Ozer  EMBrindis  CDMillstein  SGKnopf  DKIrwin  CEJ America's Adolescents: Are They Healthy?  San Francisco University of California, San Francisco, National Adolescent Health Information Center1998;
Not Available, Whatever Happened to Childhood? The Problem of Teen Pregnancy in the United States.  Washington, DC National Campaign to Prevent Teen Pregnancy1997;
Kirby  D No Easy Answers: Research Findings on Programs to Reduce Teen Pregnancy.  Washington, DC The National Campaign to Prevent Teen Pregnancy1997;
Zabin  LSEmerson  MRRingers  PASedivy  V Adolescents with negative pregnancy test results: an accessible at-risk group.  JAMA. 1996;275113- 117Google ScholarCrossref
Zabin  LSSedivy  VEmerson  MR Subsequent risk of child-bearing among adolescents with a negative pregnancy test.  Fam Plann Perspect. 1994;26212- 217Google ScholarCrossref
Furstenberg  FF  JrLevine  JABrooks-Gunn  J The children of teenage mothers: patterns of early childbearing in two generations.  Fam Plann Perspect. 1990;2254- 61Google ScholarCrossref
Grogger  JBronars  S The socioeconomic consequences of teenage childbearing: findings from a natural experiment.  Fam Plann Perspect. 1993;25156- 161174Google ScholarCrossref
Maynard  Red Kids Having Kids: A Robin Hood Foundation Special Report on the Costs of Adolescent Childbearing.  New York, NY Robin Hood Foundation1996;
Coley  RLChase-Lansdale  P Adolescent pregnancy and parenthood: recent evidence and future directions.  Am Psychol. 1998;53152- 166Google ScholarCrossref
Crockett  LJPetersen  AC Adolescent development: health risks and opportunities for health promotion. Millstein  SGPetersen  ACNightingale  EOeds. Promoting the Health of Adolescents: New Directions for the Twenty-first Century. New York, NY Oxford University Press1993;13- 37Google Scholar
Blum  RW Healthy youth development as a model for youth health promotion.  J Adolesc Health. 1998;22368- 375Google ScholarCrossref
Zoccolillo  MMeyers  JAssiter  S Conduct disorder, substance dependence, and adolescent motherhood.  Am J Orthopsychiatry. 1997;67152- 157Google ScholarCrossref
Bearinger  LHWildey  LGephart  JBlum  RW Nursing competence in adolescent health: anticipating the future needs of youth.  J Prof Nurs. 1992;880- 86Google ScholarCrossref
Blum  R Physicians' assessment of deficiencies and desire for training in adolescent care.  J Med Educ. 1987;62401- 407Google Scholar
George  M Family planning: better training for practice nurses.  Nurs Stand. September1993;722- 23Google Scholar
Blum  RWBearinger  LH Knowledge and attitudes of health professionals toward adolescent health care.  J Adolesc Health Care. 1990;11289- 294Google ScholarCrossref
Orr  DPWeiser  SPDian  DAMaurana  CA Adolescent health care: perceptions and needs of the practicing physician.  J Adolesc Health Care. 1987;8239- 245Google ScholarCrossref
Ellen  JMFranzgrote  MIrwin  CE  JrMillstein  SG Primary care physicians' screening of adolescent patients: a survey of California physicians.  J Adolesc Health. 1998;22433- 438Google ScholarCrossref
Millstein  SGIgra  VGans  J Delivery of STD/HIV preventive services to adolescents by primary care physicians.  J Adolesc Health. 1996;19249- 257Google ScholarCrossref
Tracey  JMArroll  BRichmond  DEBarham  PM The validity of general practitioners' self assessment of knowledge: cross-sectional study.  BMJ. 1997;3151426- 1428Google ScholarCrossref
Baxley  SGBrown  STPokorny  MESwanson  MS Perceived competence and actual level of knowledge of diabetes mellitus among nurses.  J Nurs Staff Dev. 1997;1393- 98Google Scholar