The United States is a large and heterogeneous country with marked differences in demography and cultural values among various states. In the United States there are marked regional differences in population density. For example, the District of Columbia, Rhode Island, and New Jersey have an average of more than 1,000 people per square mile. In contrast, Alaska, Montana, and Wyoming have fewer than 7 people per square mile. Providing medical services may be more difficult in areas of low population density.1 For many complex medical services, a critical mass of both patients and trained clinicians is necessary to make the service viable.
Providing reproductive health services for teens can be both controversial and complex. There is significant controversy over the role of teens in determining their reproductive health choices and the appropriate degree of influence that should be exerted by their parents. In addition, providing reproductive health services such as pregnancy termination for teens can be complex often requiring specialized practice sites, physicians from multiple disciplines, nurses, counselors, legal experts, and other support staff. The focus of this study is to explore the association between population density and the percentage of teenage females, aged 15 to 19 years, who conceive and the percentage of pregnant teens who electively terminate their pregnancy.
MATERIALS AND METHODS
Population data and land area in square miles for each state was obtained from the 2000 United States Census.2 Population data were rounded to the nearest 1,000 people. Geographic regions consisting of multiple adjacent states were defined by using the American College of Obstetricians and Gynecologists (ACOG) districts. Non-US regions of ACOG were excluded from each ACOG district for the purpose of this analysis. For the purpose of this analysis the geographical boundaries of the districts were District I: Maine, New Hampshire, Massachusetts, Vermont, Connecticut, and Rhode Island; District II: New York; District III: New Jersey, Delaware, and Pennsylvania; District IV: District of Columbia, West Virginia, Maryland, North Carolina, South Carolina, Georgia, Florida, and Puerto Rico; District V: Michigan, Indiana, Ohio, and Kentucky; District VI: North Dakota, South Dakota, Minnesota, Wisconsin, Illinois, Iowa, and Nebraska; District VII: Kansas, Missouri, Tennessee, Oklahoma, Arkansas, Alabama, Mississippi, Louisiana, and Texas; District VIII: Alaska, Washington, Oregon, Idaho, Montana, Wyoming, Utah, Nevada, Arizona, New Mexico, Colorado, and Hawaii; and District IX: California.
Teen population, pregnancy, and abortion data, for teens aged 15 to 19 years, in 1999 and 2000 were obtained from the Alan Guttmacher Institute,3 Tables 5 to 7 and Tables 12 to 14. As noted in the report, the number of births to teenagers in each state and teenage birthrates were obtained from the National Center for Health Statistics. The annual numbers of abortions for each state were calculated from survey data collected from abortion providers by the Alan Guttmacher Institute. The assignment of teens to state of residency was performed on the basis of information provided by state abortion agencies. The calculation method includes the assumption that teens travel outside their home state for abortion services in a similar proportion as older women. The numbers of abortions by age, race, and ethnicity were estimated from state health department data. Complete annual reports for each state were not available for each calendar year. The Pearson correlation coefficient was used to analyze relationships between variables.
The relation between population density and both teen pregnancy rate and the percentage of teen pregnancies electively terminated was analyzed by both ACOG districts and states for both 1999 and 2000.
Figure 1 displays the relationship between population density and teen pregnancy rate, for teens aged 15 to 19 years, by ACOG district for 2000. In 2000, there was no significant correlation between population density and teen pregnancy rate when analyzed by ACOG district (correlation coefficient, r = 0.063, P = .87). In 2000 there was no significant correlation between population density and teen pregnancy rate when analyzed by state (r = 0.074, P = .61). Teen pregnancy rate was similar in regions of below average and above average population density. In 2000, for teens aged 15 to 19 years, the states with teen pregnancy rates 10% or greater included Texas, Mississippi, Nevada, Arizona, and New Mexico. Four states had teen pregnancy rates less than 5%: New Hampshire, Vermont, North Dakota, and Minnesota.
Figure 2 displays the relationship between population density and the percentage of teen pregnancies that were electively terminated for each ACOG district for 2000. In 2000, there was a significant positive correlation between population density and the percentage of pregnant teens who electively terminated their pregnancy when analyzed by ACOG district (r = 0.85, P = .003). In 2000, there was a significant positive correlation between population density and the percentage of teen pregnancies that were electively terminated when analyzed by state (r = 0.69, P < .001). The percentage of pregnant teens who electively terminated their pregnancy was greater in states of high population density. In 2000 more than 50% of teen pregnancies ended in elective termination in 2 states, New York and New Jersey. In seven states, West Virginia, Kentucky, Oklahoma, Arkansas, Louisiana, Utah, and South Dakota less than 15% of teen pregnancies ended in elective termination.
When teen pregnancy data from 1999 was used for the analysis, results similar to those reported for 2000 were observed. In 1999, there was no significant correlation between population density and the teen pregnancy rate when analyzed by ACOG district (r = 0.11, P = .77). When the data were analyzed for each state, no significant correlation between population density and teen pregnancy rate was observed (r = 0.079, P = .58). In 1999, for teens aged 15 to 19 years, the states with pregnancy rates 10% or greater included Florida, Texas, Mississippi, Nevada, Arizona, New Mexico, and California. Three states had teen pregnancy rates less than 5%: New Hampshire, Vermont, and North Dakota.
In 1999, there was a significant positive correlation between population density and the percentage of pregnant teens, aged 15 to 19 years who electively terminated their pregnancy when analyzed by ACOG district (r = 0.86, P = .003). Similarly, there was a significant positive correlation between population density and the percentage of pregnant teens who electively terminated their pregnancy when analyzed by state (P = .68, P < .001). In 1999 more than 50% of teen pregnancy ended in elective termination in 2 states, New York and New Jersey. In 7 states, West Virginia, Kentucky, Oklahoma, Arkansas, Mississippi, Louisiana, and Utah less than 15% of teen pregnancies ended in elective termination.
In this study population density, analyzed both by ACOG district and state, was not significantly correlated with teen pregnancy rate, but it was significantly correlated with the percentage of teen pregnancies that were electively terminated. The factors that support this association are not clear. A very simple explanation for the observed result is that teens from geographies of low population density may prefer to give birth than to elect to terminate their pregnancy. A similar explanation is that the cultural values of families and communities in states with low population density guide teens to choose birth as a preferable option to abortion. Alternatively teens living in regions of low population density may not be able to access a full spectrum of reproductive health services as easily as teens living in high population density regions. It is likely that neither of these simple explanations fully account for the observed association. Population density is likely to be a proxy for a complex set of medical, familial, religious, economic, and legal factors that influence reproductive health choices.
In regions of low population density many factors might limit the access of pregnant teens to a full spectrum of reproductive health services. For example, there may be fewer clinicians willing to provide pregnancy termination services in areas of low population density. District leaders of ACOG might want to better understand the medical factors that influence the availability of pregnancy termination services in their locales. Most residency programs and many advanced training programs for pregnancy termination are located in areas of high population density, not in areas of low population density. Physicians often choose to practice near the site of their final medical training, and abortion providers are concentrated in urban areas4 In areas of low population density clinicians with the skills to provide abortion services might be subject to more stigmatization by colleagues and the community. This could result in a decrease in the willingness for trained clinicians to provide the service in these areas. Medical leaders might explore how to develop training programs for pregnancy termination services in areas of low population density and how to recruit physician providers to these areas.
It would be of interest to further analyze the relative transportation barriers faced by pregnant teens trying to access pregnancy termination services or obstetric services. One hypothesis is that in areas of low population density it is much more difficult for pregnant teens to travel to a site providing pregnancy termination than to a site providing obstetric services, thereby influencing their choice. If such an observation were made, it might be useful to focus on enhanced methods of delivering pregnancy termination services to women in low population density geographies by ensuring, where possible, that sites providing obstetric care also strive to offer access to pregnancy termination services.
An alternative explanation is that areas of low population density have a disproportionate number of teens and parents who favor birth over pregnancy termination compared with other more densely populated regions. Additional social science research is warranted to understand the influence of population density on the degree to which teens and their parents favor pregnancy termination or delivery should a teen become pregnant. It would be of significant interest if the study demonstrated that there was a higher degree of discordance in the choice of pregnancy termination among teens and their parents in geographic areas of low population density; with the result that the parents’ choice for birth overrides the teen's preference for pregnancy termination. Alternatively, it may be that in areas of high population density there is a high degree of discordance in the choice of pregnancy termination among teens and their parents with the result that the parents’ choice for pregnancy termination overrides the teen's preference for birth. Factors such as race, ethnicity, religious beliefs, education level, and family income may influence these important choices.
Providing reproductive health care services to teens is both controversial and complex. A major controversy is the degree to which the reproductive health choices of teens should be influenced by their parents. Pregnancy termination procedures require completion of informed consent documents. In most states, many teens are not empowered to individually consent to pregnancy termination without the support of their parents or a judge. Some authorities contend that the reproductive choices of teens should rest exclusively with their parents. Other authorities believe that pregnant teens should have the freedom to make their own reproductive choices independent of their parents. In some states, the judiciary is active in providing teens parental waivers to make important reproductive health choices. In other states the judiciary is less active and less accessible in supporting the reproductive health choices of teens. Based on the findings of this study, it would be interesting to explore the degree to which population density influences the activism of the judiciary in helping teens access pregnancy termination services.
Prevention of teen pregnancy would significantly improve the health of adolescents. In this study there was no significant correlation between teen pregnancy rate and population density within the 9 ACOG districts. It is interesting that areas of high population density, where there may be a greater concentration of reproductive health services, had a teen pregnancy rate that was similar to areas of low population density. Much more needs to be done to reduce the rate of teen pregnancy in the United States. By age 18, more than 60% of teens are sexually active, and the majority of teen pregnancies are unintended.5 Between 1988 and 2000 teenage pregnancy rates decreased in every state. This decrease is probably the result of successful implementation of multidimensional and comprehensive contraception programs that recognize the unique needs of each teen.6 It is estimated that most of the decline in teen pregnancy has been due to more effective use of contraception and a small portion of the decline has been due to an increase in abstinence among teens.7 Consequently the access of teens to contraception and their motivation to use contraception are key factors in preventing unwanted teen pregnancies.8 The access of teens to reproductive health services is influenced by public policy governing issues of parental notification, support for school based clinics, content of sex education curricula in schools, and the degree of reliance on abstinence as a one-dimensional approach to teen pregnancy.9,10 Enhancing the access of teens to comprehensive reproductive health services is important to their health.