The extent to which parents should be involved in their adolescent children's decisions to have sex and use contraceptives is a sensitive issue, as well as one that has policy and programmatic implications. Positive parent-child relations are associated with fewer risk-taking behaviors, delayed sexual activity and fewer births among adolescents.¹ Many adolescent females can access confidential family planning and STD services, but recent legislative efforts have sought to require parental involvement for minors seeking family planning services at publicly funded clinics.² Policymakers and advocates who support confidential reproductive health care for minors contend that such access is essential because some sexually experienced adolescents would avoid seeking contraceptive or STD services if they were required to involve their parents. To date, 21 states and the District of Columbia have enacted laws that explicitly allow adolescents younger than 18 to consent to contraceptive services.³ No states require parental consent or notification for all minors seeking contraceptive services, though Utah and Texas require parental consent for all adolescents seeking family planning services at clinics that receive certain types of state funding.

Most family planning clinics place priority on providing confidential services to adolescents without parental knowledge or approval, but recognize that parents play a central role in many adolescents' lives. Consequently, many have made efforts to facilitate and improve parental involvement in adolescents' lives. For example, among federally funded family planning agencies, which served 1.9 million adolescents in 2001, 43% had programs or policies intended to enhance parent-child communication.⁴

Almost two-thirds of 15–17-year-old females who obtained a family planning or medical service in the past year did so from clinics.⁵ However, little is known about relations between adolescents who use clinic-based sexual health services and their partents. One purpose of this article is to describe communication, connectedness with parents and perceived parental attitudes among females younger than 18 visiting family planning clinics. Furthermore, because most such clients report that parents know they use the clinic for sexual health services,⁶ this article also examines associations between adolescents' relationships with parents and parental knowledge of clinic visits.

RESEARCH ON PARENT-CHILD RELATIONS

A substantial body of research has examined parental relationships among the general adolescent population. Three of the more relevant indicators for purposes of this study are levels of parent-child communication about sexuality issues, parent-child connectedness and parental attitudes toward sex and contraception.

Communication About Sex and Contraception

Over the last two decades, researchers have devoted a significant amount of attention to the extent and quality of parent-child communication about sex. Most recently, the 2002 National Survey of Family Growth (NSFG), which gathered information from a representative sample of women (and men) of reproductive age, found that a majority of females aged 15–17 have talked to parents about issues such as how to say no to sex (62%), methods of birth control (52%) and STDs (57%); substantial minorities have talked to parents about where to get birth control (39%) and how to use a condom (30%). Twenty-five percent have not talked to a parent about any of these issues.⁷

Likewise, other studies have consistently found that about half of teenagers have talked to parents about sex and birth control,⁸ but the current state of communication is hardly ideal—many adolescents express a desire for more information and guidance from parents than they receive on these issues.⁹ Also, studies with information from both parents and their adolescent children have found that mothers are more likely to report that they have discussed a given topic than are their children,¹⁰ suggesting that parents are not conveying information as effectively as they think they are.

Parent-Child Connectedness

Parent-child connectedness, defined as "an emotional bond between a parent and a child that is both mutual and sustained over time,"¹¹ has received increasing attention among researchers, policymakers and the popular press. This is partly because research using the National Longitudinal Study of Adolescent Health (Add Health), a representative survey of adolescents in school and their parents, has consistently found that connectedness to parents is associated with a decreased likelihood of sexual intercourse and risky adolescent behaviors, including pregnancy.¹² Connectedness with parents is typically measured as a composite of several relationship dimensions, such as closeness, satisfaction with the overall relationship, satisfaction with communication, and perceptions of caring and feeling loved and wanted by family members.¹³ Most adolescents report high levels of connectedness with parents. A scale measure of mother-child connectedness using data from Add Health ranged from three to 22; the mean score of 19 suggested that on average, a majority of U.S teenagers are highly connected with parents.¹⁴ Similarly, research using a scale to measure parental responsiveness among a sample of high-risk minority adolescents suggests that a substantial proportion of this population also are connected to parents (mean, 22; scale range, 8–32).¹⁵

Parental Attitudes Toward Sex and Birth Control

Parents often disapprove of their adolescent children's having sex. One Add Health study found that 87% of mothers strongly disapprove of their daughters' having sex;¹⁶ another study, of high-risk teenagers, found 75% of mothers disapproving of sex.¹⁷

Just as adolescents are less likely than their mothers to recall previous conversations about sex, they perceive mothers to be less disapproving than they actually are. While 61% of eighth and ninth graders (males and females) in one study perceived strong maternal disapproval of their having sex, 76% of mothers expressed such disapproval.¹⁸ A similar pattern has also been found among a sample of high-risk teenagers.¹⁹

Parental attitudes about contraception are not as well documented. Only 21% of mothers in Add Health report that they have recommended a birth control method to their minor daughters,²⁰ but lack of recommendations does not necessarily mean disapproval. At least one study found that a majority of parents do not object to their adolescent children's having access to contraception. Almost two-thirds of Baltimore parents registering their children for services at middle and high school–based health clinics indicated that the clinics could, in the future, provide adolescents, and presumably their own children, with access to condoms and birth control.²¹

ADOLESCENTS ATTENDING FAMILY PLANNING CLINICS

Research findings on parent-child connectedness and communication about sex among the general population may not be applicable to minor clients of family planning clinics. However, a few studies have gathered information from this population.

Most studies have found that a majority of minor adolescents who use family planning clinics report that a parent knows they are at the clinic.²² Reports of an informed parent are more common among adolescents who are younger,²³ who are black²⁴ and who have made prior contraceptive visits,²⁵ and less common among teenagers with college-educated mothers and those who live with both parents.²⁶

More than one-third of young women report that they voluntarily informed parents that they used family planning clinic services.²⁷ However, even teenagers who voluntarily tell their parents of clinic visits may not be discussing broader issues of sexuality with them. A 1984 study found that even though parental awareness of daughters' clinic visits increased over a 15-month period, the proportion of adolescent females who indicated that they "usually" talked about sex and birth control with their mother remained at 37–39%.²⁸

For a substantial minority of adolescents who use family planning services, parents are supportive of and actively involved in this aspect of their lives. In various studies, 6–15% of teenagers said they had learned about the clinic from a parent,²⁹ 24% of young women indicated that a parent had suggested the clinic for sexual health services,³⁰ and 10–14% of adolescents who made contraceptive visits were accompanied by a parent, typically the mother.³¹ Several studies suggest that black teenagers and younger teenagers are significantly more likely than whites and older adolescents to have parents involved in this aspect of their lives.³² Parental involvement presumably indicates approval of daughters' contraceptive use, but lack of overt involvement does not necessarily signify disapproval.

A number of research questions about relationships between adolescents attending family planning clinics and their parents remain unanswered. For example, to what extent do adolescent female clinic clients talk to their parents about sex and contraception? What level of connectedness do they report? Are young women seeking sexual health services with parental knowledge and support? Are parent-child connectedness and discussions about sex and birth control associated with whether young women voluntarily inform their parents that they are using family planning services? And which groups of young women who use clinics are the most likely to do so at a parent's suggestion? Our study answers these questions using current, national data from more than 1,500 females younger than 18 obtaining sexual health services in U.S. family planning clinics.

METHODS

Data

We systemically selected a random sample of clinics from a database of all publicly funded family planning clinics in the United States maintained by the Guttmacher Institute. The universe was restricted to facilities that served 200 or more adolescent contraceptive clients in 2001, to maximize the cost-effectiveness of fielding the survey. We stratified the universe from which the sample was drawn by adolescent caseload (200–399; 400–749; 750–1,199; 1,200 or more), type of facility (hospital clinic, health department, Planned Parenthood or other), receipt of Title X funding and whether state law explicitly protects minors' access to contraceptive services. Facilities in Texas and Utah were excluded because at least some minors in state-funded family planning clinics in those states were already required to obtain parental consent for family planning services. The final universe of 2,442 family planning clinics from which the sample was drawn represented 37% of all facilities but served 82% of all adolescent contraceptive clinic clients in 2001.

Participating facilities distributed questionnaires to eligible adolescents for 2–6 weeks. If a facility declined to participate or did not obtain usable questionnaires from at least 50% of eligible teenagers seen during the study period, it was replaced by the next clinic in the stratified sample; this procedure ensured that the replacement clinic was similar to the original. Our goal was to gather surveys from minors in 80 clinics. A total of 97 clinics declined to participate, mostly because of understaffing, time and budget constraints or general lack of interest. An additional 31 clinics agreed to participate, but failed to obtain usable surveys from at least 50% of eligible female clients. Low response rates typically stemmed from administrative disorganization, inability of facility staff to keep a complete tally of eligible minors or failure to pass out the survey. Because most clinic refusals and failures were caused by administrative issues, we expect that the adolescent clients in clinics that declined to participate or that were unable to do so successfully did not differ from the adolescents in our sample. Fieldwork began in May 2003 and lasted until February 2004. The final sample of 79 facilities consisted of 28 health department sites, seven hospital clinics, 31 Planned Parenthood affiliates and 13 other clinic types. Clinics from 33 states participated, including several from each of the four major census regions.

Data were collected via a self-administered questionnaire, available in both English and Spanish. The questionnaire was written at a fifth-grade reading level, and results from pretesting indicated it took approximately 10 minutes to complete. Clinic staff distributed surveys to all clients younger than 18 seeking sexual health services, excluding abortion and prenatal and postnatal care. Surveys were anonymous and confidential, and respondents were given a sealable envelope so that clinic staff could not view their completed questionnaires. The survey instrument and fielding protocol were approved by the Guttmacher Institute's institutional review board as well as external institutional review boards of several study clinics that required separate approval.

Participating clinics saw 2,038 eligible female clients during the survey period, and obtained surveys from 75%. The final data used for these analyses come from 1,526 young women. We constructed weights to represent the universe of adolescent clients using all publicly funded family planning clinics that serve 200 or more adolescents annually. We imputed missing values for key demographic variables using a hot-deck procedure.³³

Measures

•Communication about sexual health issues. Our measure is based on a question that asked respondents how often they discussed seven topics with a parent: how to say no to sex, how to prevent getting an STD, how to prevent getting HIV/AIDS, how to prevent pregnancy, how to use condoms, how to know when it is appropriate to have sex and where to get prescription contraceptives. Respondents were provided with four response categories, ranging from "a great deal" to "not at all." The seven items were highly interrelated (Cronbach's alpha, 0.92) and were used to construct an additive scale (range, 7–28), on which a higher score indicated more frequent conversations on more topics.* We collapsed the scale into a four-category ordinal variable, with each category containing approximately 25% of respondents, to make discussion of the findings more meaningful and straightforward.

•Parent-child connectedness. Our measure of parent-child connectedness is based on responses from two sets of items, both adopted with minor modification from Add Health. The first assessed communication with parents ("I am satisfied with the way my mother/female guardian and I communicate with each other"); the second assessed overall relationships ("Overall, I am satisfied with my relationship with my mother/female guardian"). Parallel items about communication and overall relationship with fathers were also asked. There were five response categories, ranging from "strongly agree" to "strongly disagree," plus "not applicable." For the analyses, we collapsed the response categories of the items so that "strongly agree" and "agree" were considered one response, as were "strongly disagree" and "disagree."

We first constructed a single measure of satisfaction with communication with parents and a single measure of overall relationship with parents. Since approximately one-half of minor females did not live with fathers, we relied primarily on information pertaining to mothers, but used information pertaining to fathers when maternal information was missing. The two items were highly related (Cronbach's alpha, 0.92), and were combined to construct a three-category measure of connectedness with parents: satisfied with both communication and overall relationship, satisfied with only one dimension (or neutral on both) and dissatisfied with both. This measure is simpler than measures of parent-child connectedness used in prior research,³⁴ but it captures two of the more important dimensions of this construct.

•Parental attitudes toward sex and contraception. Items assessing adolescents’ perceptions of parental attitudes also were adopted with minor modification from Add Health: "How would your mother/female guardian feel about your having sex at this time in your life?" and "How would your mother/female guardian feel about your using birth control at this time in your life (if you were having sex)?" We also asked parallel questions about fathers’ attitudes. There were five response categories, ranging from "strongly approve" to "strongly disapprove," plus "don’t know." For the analyses, we collapsed "strongly approve" and "approve" into a single category and did the same with "strongly disapprove" and "disapprove." We created one variable for sex attitude and one variable for birth control attitude. Again, we relied primarily on information about maternal attitudes, and used information about fathers’ attitudes when information about mothers’ attitudes was not provided. We combined our measures of perceived parental attitudes toward sex and toward birth control to construct a four-category measure of parental attitudes: approves of contraception and approves of or is neutral about sex, approves of contraception but disapproves of sex, disapproves of both and a residual category that included other responses (including "don’t know"). (We examined other combinations of these two variables, and determined that this categorization scheme was the most efficient and provided the most insights into how perceived parental attitudes are associated with parental knowledge of clinic visits.)

•Parental knowledge of clinic visit. Respondents were asked "Does a parent/legal guardian know you come to this clinic for birth control or other sexual health services?" Teenagers who answered yes were directed to a follow-up item asking if they had voluntarily informed their parents, if their parents had recommended the clinic or if their parents had found out through some other means—for example, through an insurance company or another person. We constructed three measures from these questions. We first examined whether young women indicated a parent knew they were at the clinic. Among those whose parents knew, we then examined whether young women had voluntarily informed parents and whether parents had suggested the clinic visit.

•Demographic characteristics. We asked respondents about their age, race and ethnicity, mother’s education, living arrangement, number of prior births and, when appropriate, number of months since first intercourse and number of contraceptive visits in the last 12 months.

•Missing values. The overwhelming majority of respondents provided usable responses to most items. For example, only 2% did not provide their month and year of birth, one of the first items on the survey, and 6% failed to provide information about mothers’ education, one of the last items. Nonresponse and missing values for items pertaining to parents’ sex and birth control attitudes were higher than those for other items. When asked how their mothers would feel about their having sex, 7% of young women indicated "don’t know"; 3% did not answer the item. Some 75% of these respondents were living with a mother; we hypothesize that a response of "don’t know" in this circumstance indicated a lack of communication about these issues and, hence, the inability to provide a response. Therefore, we incorporated "don’t know" as a usable response category in assessing young women’s perceptions of parental attitudes.

Analytic Strategy

All analyses use weighted data. We first provide a demographic profile of women younger than 18 attending family planning clinics and, to the extent possible, compare them with the general population of females aged 15–17, as well as those who are sexually experienced, using nationally representative data from the 2002 NSFG. (We do not assess whether differences between the populations are statistically significant.) We next examine descriptive statistics to assess the extent to which young women attending family planning clinics talk to their parents about sex and contraception, levels of connectedness with parents and perceptions of parents' attitudes toward sex and birth control. The next step is to use t tests to assess bivariate associations between these three dimensions of teenagers' relationships with parents and our measures of parental knowledge of clinic visits. Our final step, using logistic regression, is to determine if these associations are maintained once demographic characteristics and sexual behavior are taken into account. Stata software, version 8, is used to conduct tests of significance that take into account the clustered nature of the sample of clients within clinics.

Given the cross-sectional nature of the data, we cannot determine if communication with parents, parent-child connectedness and perceived parental attitudes preceded parental knowledge of clinic visits or had any influence on how parents knew of the visits. Hence, our analyses are more descriptive than explanatory.

RESULTS

Characteristics of Minors Attending Clinics

Nine percent of minor females who were attending family planning clinics in the United States in 2003 were younger than 15, 18% were 15 years old, 31% were 16 and 42% were 17 (Table 1).^† The majority (56%) were white, 23% were black, 15% were Hispanic and 7% were of some other race or ethnicity. Twenty-one percent of respondents had mothers who were not high school graduates; 20% had mothers who were college-educated. Similar proportions of young women resided with both parents and with a single mother (43% and 44%, respectively). Respondents who were sexually experienced (91%) had been so for an average of 20 months (not shown). Only 5% had given birth, and 58% had made one or more contraceptive visits in the last 12 months.

Adolescents using family planning clinics differed from adolescents nationally (in 2002) on a number of indicators. Black adolescents made up only 16% of the population aged 15–17, and were overrepresented among clinic clients (23%—Table 1).³⁵ White adolescents were underrepresented among the clinic population. Minor adolescents using family planning clinics are likely to be somewhat economically and socially disadvantaged relative to the larger population of minor adolescents, given that greater proportions of them are black, have mothers with a high school education or less, and do not live with both parents—all characteristics associated with economic disadvantage.

While sexually experienced adolescents using family planning clinics for reproductive health services were demographically similar to the larger population of sexually experienced minor adolescent females, there were some differences. A lower proportion of minors using family planning clinics than of sexually experienced adolescents nationally were white, and a greater proportion had mothers who had not graduated from high school; there is some indication that living in a two-parent household was more common among clinic clients. Also, a lower proportion of clinic clients had given birth (though this may be because the clinic sample included adolescents younger than 15 as well as teenagers who were not sexually active).

Relations with Parents

Most respondents had talked with a parent about sex and contraception (Table 2). The largest proportions had talked a great deal with parents about how to say no to sex (42%), and had discussed how to prevent getting an STD, how to prevent HIV/AIDS and where to get prescription birth control (32–33%). Fifty percent of young women indicated they did not talk with parents at all about how to use condoms, making it the least commonly discussed of the seven issues. The mean score on the scale that combines all seven items was 17.7, about the midpoint (not shown), suggesting that on average, teenagers using family planning clinics have moderate levels of discussion with parents about these issues. Most teenagers had talked about these topics with parents at least a little bit (50–80%), and only 7% had not talked about any of the seven issues with parents (not shown). However, when asked to provide an overall evaluation of these conversations, 29% indicated they were very helpful, and 22% indicated that they were not at all helpful or did not occur.

In general, young women reported positive relations with at least one parent (Table 3). Approximately three-quarters were satisfied with their overall communication and overall relationship with their mothers, and one in 10 strongly disagreed with either statement (not shown). Young women's evaluation of relationships and communication with fathers was not as positive. Fewer than half were satisfied with communication with their fathers, and slightly more than half were satisfied with the overall relationship.

Our measure combining information on relationships with both mothers and fathers on communication and overall relationship shows that 68% of young women were highly connected with parents (Table 4, page 198). Some 14% were disconnected, insofar as they were dissatisfied on both dimensions.

Young women largely perceived parents to disapprove of their being or becoming sexually active, but felt that parents were more supportive of their using contraceptives (Table 3). More than one-half of young women reported that their mothers disapproved of their having sex, but nearly two-thirds perceived that if they were going to have sex, their mothers approved of their using birth control. Fathers were perceived as having far more negative attitudes; 61% of young women indicated that fathers disapproved of their having sex, and only 36% perceived paternal approval of birth control.

Combining information from both items on parental attitudes, we found that young women using family planning clinics most commonly perceived parents to approve of birth control and either approve of or be neutral about their having sex (34%—Table 4). Thirty percent perceived parents to approve of their using contraceptives (if they were sexually active) but to disapprove of their having sex. Nineteen percent perceived parents to disapprove of both behaviors.

Parents' Knowledge of Clinic Visits

Sixty percent of young women indicated that a parent knew they were at the clinic for sexual health services (Table 5, page 198). Among these, 66% had voluntarily told parents, and 40% were there at a parent's recommendation. Nine percent indicated that a parent had found out some other way (not shown).^‡ The three responses were not mutually exclusive, and 13% of those with informed parents indicated that their parents knew of the visits through multiple mechanisms. The most common multiple response (given by 10% of those with informed parents) was that parents had suggested the clinic and young women had told their parents voluntarily. In these instances, parents may have recommended the clinic for future health care needs, and upon subsequently accessing services, young women had informed parents.

The bivariate associations suggest that parental relations are associated with whether parents know their daughters are at family planning clinics, but are less consistently associated with how parents learned of the visits. The more young women communicated with parents about sexuality issues, the more likely they were to indicate that a parent knew they were at the clinic. However, among teenagers with informed parents, there was no significant association between level of communication and how parents knew they were at the clinic.

Among adolescents who were most connected with their parents, 71% indicated that a parent knew they were at the clinic; this proportion was significantly higher than the 37% for teenagers who exhibited the lowest levels of connectedness. Notably, 50% of young women whose parents did not know they were at the clinic reported high levels of connectedness (not shown).

Adolescents' perceptions of parental attitudes toward sex and contraception were associated with several aspects of parental knowledge of their clinic visits. Among young women who perceived parents to disapprove of their both having sex and using contraceptives, only 25% indicated a parent knew they were at the clinic; the proportion was significantly higher for every other attitudinal group (49–77%). Some 44% who perceived parents to have disapproving attitudes had voluntarily told parents they were at the clinic, compared with 61–73% for the other attitudinal groups.

Many of the associations between parent-child relations and parental knowledge of clinic visits were maintained even after demographic characteristics were controlled for (Table 6). Among all minors using family planning clinics, discussions about sexuality were strongly associated with parental knowledge of clinic use. Those who reported the highest level of communication had five times the odds of indicating that a parent knew they were at the clinic relative to young women who had the lowest levels. Female adolescents who were most connected with parents had two times the odds of the least connected of indicating that a parent knew they were at the clinic. Minor adolescents who perceived parents to approve of birth control were significantly more likely to indicate that a parent knew they were at the clinic than were young women whose parents disapproved of both behaviors (odds ratios, 5.4 and 4.5 depending on perceived parental attitude toward sex).

Associations between demographic and social characteristics and parental knowledge of the clinic visit confirmed findings from prior research, suggesting that they operate independently of parent-child relations.³⁶ Relative to 17-year-olds, younger teenagers were more likely to indicate that a parent knew they were at the clinic; the differential was particularly marked for those aged 14 and younger (odds ratio, 3.7). Relative to young women with college-educated mothers, those whose mothers had no more than a high school education were more likely to indicate that a parent knew they were at the clinic. Young women who lived only with their mothers were more likely than those living with two parents to report that a parent knew they attended the clinic. Finally, young women who had made one or no visits for contraception in the past year were less likely to have parents who knew they were at the clinic, and the longer the time since first sex, the more likely were parents to know about the clinic visit.

Among young women whose parents knew they were at the clinic, a second logistic regression model compared those whose parents suggested the clinic with those who voluntarily informed parents;^§ respondents who indicated their parents had found out some other way were excluded from the model. Few of the parental relationship dynamics were associated with how parents knew. The likelihood that a parent had recommended the clinic as opposed to the adolescent's voluntarily informing parents did not differ by level of connectedness with parents or communication about sexuality issues. Relative to adolescents whose parents disapproved of sex and birth control, only those in the "residual" group (youth who gave inconsistent answers or answers that did not seem meaningful) were significantly less likely to indicate a parent had recommended the clinic (and were more likely to have told parents voluntarily). Because the three measures of parental relations are somewhat interrelated among adolescents with informed parents, we examined models that included each relationship dimension separately and controlled only for demographic characteristics (not shown). Even in those analyses, communication and connectedness were not associated with parent's suggesting the clinic visit.

Finally, several demographic characteristics and sexual behaviors were associated with whether a parent had recommended the clinic. Minor females aged 15 and younger were more likely than 17-year-olds to indicate that a parent had recommended the clinic. Adolescent females who lived with their fathers but not their mothers were less likely than those who lived with both parents to report that a parent had suggested the clinic; those who had had a prior birth were less likely to be there at a parent's suggestion than those who had not given birth. The longer minor female adolescents had been having sex, the less likely they were to indicate that a parent had recommended the clinic.

DISCUSSION

The majority of minor adolescent females attending family planning clinics have good relationships with at least one parent. Most have talked to a parent about a range of issues related to sex and birth control, and most report high levels of connectedness. A majority of young women perceive parents to disapprove of their having sex, but two-thirds perceive that their parents approve of their using contraceptives if they are sexually active. This seemingly inconsistent perception mirrors findings from a public opinion poll indicating that most adults do not approve of teenagers' having sex but support teenagers' having access to contraceptives if they are going to have sex.³⁷

It is important not to overlook that one in seven adolescents were "disconnected" from parents. These young females may need support in making responsible decisions about their sexual health.

A majority of minor females indicated that a parent knew they were at the clinic. Young women who had the highest level of communication with their parents about sexual matters, adolescents who were highly connected with parents and those who did not perceive parents to disapprove of sex and contraception were more likely to indicate that a parent knew they were there than those with low levels of communication and connectedness, and those whose parents disapproved of both sex and birth control. Notably, we were unable to establish the causal direction of these relationships, and there may be several paths. For some teenagers, high levels of communication may have preceded the clinic visit; for example, discussions about how to prevent pregnancy and where to get birth control may have motivated teenagers to visit the clinic. Alternately, or in addition, parents may have initiated more discussions about sexual matters such as birth control and STDs after their daughters disclosed that they were obtaining services.

Adolescents using family planning clinics are more likely than U.S. teenagers in general to be economically and socially disadvantaged, though they are demographically similar to the national population of sexually experienced minor teenagers. Interestingly, minors from relatively better-off households—for example, those who lived with both parents or whose mothers had college degrees—had reduced odds of indicating that a parent knew they were at the clinic. These adolescents may use family planning clinics, as opposed to private physicians, specifically because they are concerned about confidentiality; future research should explore this issue in more depth.

Parents may learn about their daughters' use of family planning clinics through several mechanisms. Most commonly, adolescents had voluntarily told their parents, but a substantial minority of parents had themselves suggested that their daughters use the clinic. Among young women whose parents knew they were at the clinic, neither communication about sexual issues nor connectedness was associated with whether parents had recommended the clinic.

Expanding on previous research showing that younger adolescents are more likely than older adolescents to report that a parent knows they are at the clinic,³⁸ we found that those aged 15 and younger were also more likely than older adolescents to be at the clinic at a parent's suggestion. Barriers such as lack of information and misinformation about contraceptives and where to obtain them,³⁹ limited mobility and restricted financial independence are more pronounced for the youngest adolescents than for older teenagers. Our research suggests that many adolescents aged 15 and younger access family planning services specifically because parents are involved.

Four in 10 adolescent females indicated that a parent did not know they were at the clinic for sexual health services; the proportion was higher among adolescents whose parents disapproved of sex and birth control than among those reporting any other parental attitude. This pattern suggests that some teenagers perceive that they are using sexual health services against parents' wishes. Prior research has found that teenagers whose parents do not know they are at the clinic would be particularly likely to engage in unsafe sex if parental involvement were mandated; the most common reasons adolescents do not tell parents they use family planning clinics are that they do not want parents to know they are having sex (25%) and they are concerned that parents will be disappointed if they find out they are having sex (22%).⁴⁰ Given that one-half of teenagers whose parents did not know they were at the clinic were highly connected with parents, we expect that these teenagers would opt to have unsafe sex rather than disappoint parents and jeopardize their good relations.

Our study has several shortcomings. Even though participating clinics represent a cross section of U.S. family planning providers with large adolescent caseloads, the high number of clinic refusals and survey administration failures may have resulted in a biased sample, meaning that the findings cannot be extrapolated to all minor females using such clinics. The reading level and length of the survey could have deterred some teenagers from participating or caused them to answer items incorrectly. As is the case among the larger population of adolescents, minor females using family planning clinics may have inaccurate perceptions of parents' attitudes toward sex and birth control. Biased responses are possible if teenagers falsely reported that parents were aware of the visit in order to preserve their access to confidential services. While we refer to our measures as assessments of parent-child relations, the measure we constructed gave primary emphasis to relationships with mothers and incorporated teenagers' evaluations of their relationships with fathers only when information from mothers was not available. This strategy was warranted because a substantial minority of minor adolescent females are not in contact with their fathers. However, relationships with fathers are sometimes associated with young women's reproductive health outcomes,⁴¹ and subsequent research should develop measures that incorporate relationships with both fathers and mothers in ways that do not penalize, or assign lower scores, to young women who live only with their mothers.

Family planning clinic visits provide appropriate opportunities for clinic staff to encourage adolescents to talk to parents about sexual matters, including birth control and prevention of STDs. And a number of clinics have adopted practices and policies to promote parental involvement,⁴² typically without compromising adolescent confidentiality.⁴³ Most young women using family planning clinics have good relations with parents. Rather than mandating parental involvement for minors, and potentially jeopardizing parental relationships for the minority of adolescents whose parents do not know they are at the clinic, a more effective strategy for promoting positive parental engagement would be to increase support for activities that clinics have undertaken to improve and maintain these relations.