EXPERIMENTAL SOCIAL WORK SERVICES FOR UNMARRIED TEENAGE MOTHERS

 
 G. J. Loury, Ph.D.
 POB 880648
 Boca Raton, Florida 33488-0648

 EXPERIMENTAL SOCIAL WORK SERVICES FOR UNMARRIED TEENAGE MOTHERS
 ABSTRACT
 An experiment was conducted to increase the quality of life and reduce the number of repeat out-of-wedlock births among teenage mothers.  Teenage referrals to a longstanding traditional maternity home and clinic for unwed mothers were randomly assigned to a customary care condition and an intensive service social work experimental condition.  Clients were followed for two years after the termination of services.  Experimental clients showed large gains over controls in nearly every outcome measure.  Their number of repeat out-of-wedlock births was only one half of the control's.
 EXPERIMENTAL SOCIAL WORK SERVICES FOR UNMARRIED TEENAGE MOTHERS
 Until the 1970s, social services for unmarried mothers were motivated by the desire to redeem wayward girls, to protect the child, to conceal the mother's shame, and to prevent her from repeating her mistake (Shlakman 1966; Carabiner and Gross 1982).  Since then, although the rate of illegitimacy has remained high, the moral stigma associated with out-of-wedlock birth has apparently declined (Gallagher 1973; Ways and Means Committee 2002;  Garfinkel and McClanahan 1994).  This change in attitude can probably be attributed to a variety of factors, among them the human rights movement, an increased openness on the part of the media in presenting sexual material, and the diminished status of the traditional marriage.  In addition, many have noted that increased sexual activity among adolescents and the legalization of abortion may well have affected societal attitudes (Gallagher 1973; Ways and Means Committee 2002; Wu 1996; Maynard 1995).
 Two noteworthy trends appear to be related to the changes in attitude -- an impressive increase in the number of adolescent unmarried mothers who keep their babies and a dramatic decline in adoption rates (Friedman 1975;  Haring 1975; Ways and Means Committee 2002).  Although some pregnant unmarried women do decide to marry, to undergo an abortion, or to release their baby for adoption, it is clear that many young women prefer to establish single-parent families.  Unfortunately, single parent homes, especially those headed by never-married mothers are at high risk for a number of social problems:  poverty, drug and alcohol abuse, child abuse, crime, and, most notably additional out-of-wedlock birth.  These two trends, together with the precarious situation of never married adolescent mothers, are of major consequence to social service agencies.
 In light of the changing needs of this population of unwed mothers, agencies may be moved to review their programs to insure that the services they offer coincide with what these clients actually want and need.  Prior to developing a contemporary definition of service delivery, agencies must define specific areas of need.  In order to assess the strengths and weaknesses of these adolescents, the girls' approaches to their current living situations -- including such areas as education, employment, interpersonal relationships, and parenting -- must be taken into account.  Therefore, the present study was designed to test the value of intensive case-management and social services both in reducing the likelihood of additional out-of-wedlock births among unwed mothers and in improving the conditions of their lives:  employment, living arrangements, life satisfaction, interpersonal relationships, and child care.
 PROCEDURES
 The experiment was conducted among the clients of a prominent and longstanding home for unwed women (the Agency) that also carries a large outpatient clientel.  In 1991, 129 unmarried pregnant or recently delivered women were referred to the Agency by a variety of sources.  Amenable clients were randomly assigned to an experimental group that received intensive case-management, psychotherapy, counseling, and many referral services in addition to the Agency's customary care and to a control group that received the Agency's customary package of care -- a small amount of counseling, skills training, basic compensatory education, and referral services.
 The intensive case-management services were modeled after traditional intensive casework care augmented with a battery of additional services that would assure that each client received a large additional amount of service in comparison with customary care.  The intensive casework condition of this experiment recalls many of social work's earlier intensive casework interventions (Stein and Gambrill 1977) and the work training social service packages of the MDRC demonstrations (Gueron 1990).  It also reflects the family preservation strategies of recent years except that care in this study was not short term but was provided for a full year (Fraser, Pecora and Haapala 1991).  Most notably, the research replicated similar recent experiments but with a much more intensive array of service (Polit, Quint, and Riccio 1988; Quint, Bos, and Polit 1997; Stahler and DuCette 1991)  Intensive care was limited to one year with a provision for some on demand follow-up services arranged through the primary case-manager during the succeeding two years.   All clients who met the following criteria were included in the study:  (1) only one pregnancy had gone to term (two clients had twins), (2)  the woman must not have been married as of the time of the study, and (3) pregnancy must have occurred after the age of 14 but before age 20.  Of the 129 women referred to the Agency, 91 met the criteria.  Of these 91, 82 agreed to participate in the study.  Forty-one were randomized to the experimental group and 41 to the control.
 A multifaceted questionnaire was developed to include items about the mothers' current living situation, her educational level and employment status, and her relationship with the child's biological father and others, life satisfaction, child care arrangements, and most importantly, whether she was pregnant and if so whether she expected or did go to term.  These items were adapted from a number of existing instruments (Polit, Quint and Riccio 1988; Rancifer and Jones, in review).  All of the information relied upon client self-reports.  No independent verification of information was possible.
 Individual appointments were made with each unmarried mother in her own home or in another location of her choice.  Each of the interviews lasted approximately one hour.  Clients were periodically reinterviewed throughout the year of service and at various times during the two year follow-up.

 -- TABLE 1 ABOUT HERE --
     Table 1 compares the principal characteristics of the experimental group and the control.  The two groups did not differ significantly on any measure.  Indeed, in all measures they are virtually identical.  As expected, the two groups contain adolescents who are marginal, working little, largely out of school, and with no great amount of life satisfaction.  They are also dependent and poor.  Moreover, only 15.2% of the total sample had completed high school, while only one client had some college.
 Because of her age, lack of working experience and, in most cases, her limited educational background, the adolescent who keeps her child is in a particularly insecure position.  The greatest number of women in this study reported that their major source of income was from public assistance programs.  Minimal additional sources of income were provided by the child's biological father and parents.
 – TABLES 2 AND 3 ABOUT HERE --
 Table 2 compares the percent of each group that received different services.  It is apparent that experimental clients received far more services than control clients.  Table 3 compares the intensity of services measured by the average number of hours of services received by the experimental group and the control.  The data were collected from Agency files and the files of all agencies to which clients were referred.  Moreover, experimental subjects received an average of seven different services that lasted for 11.1 months -- almost the entirety of the experiment.  In contrast, subjects in the control received only an average of three services for only 2.6 months.  Total average hours of care in the experimental group reached 430.6 hours during the year while control clients only absorbed an average of 108.3 hours.  This difference is all the more remarkable for the fact that educational services such as counseling accounted for fully 47.4 hours of service in the control.
 Many of the women reported having friends, the majority of whom they had met after the pregnancy.  However approximately %40 felt isolated and lonely.  Although most of the respondents denied any differences in their relationships with friends from before the pregnancy, and those made subsequently, the 53.2 percent who did suggest a difference, provided a variety of distinctions.  About half the women indicated that they were closer to their old friends.
 Approximately 60 percent of the women reported that they were dating actively and most of them said they dated only one man.  Over half the women who dated were seeing a man whom they had known prior to the pregnancy.  Only three reported that they were dating their child's biological father.
 Part of the dating relationship for approximately two-thirds of the women who dated included sexual intercourse, which they typically described as enjoyable.  Those women who denied having intercourse at the time of the study suggested that it was because they did not know a man with whom they wished to share that experience.
 The possibility exists that one effect of an unplanned pregnancy is the woman's change in attitude toward sex.  Fifty-three percent of the women reported having undergone such a change, half of whom observed that they had become fearful or resentful about sexual relationships.  The remainder of the women who reported a change in attitude said they had become more responsible.  Although a variety of attitudinal changes were reported, almost all the women described a dramatic increase in the use of contraceptives following the birth.  Sixty-seven percent of them were using the pill.  Only five woman reported that neither she nor her partner used any contraceptive method at all.  From the reports it was clear that none of the women's partners used any form of contraception, which suggests that in this sample contraceptive responsibility rested on the women exclusively.
 At the time of the study, almost two thirds of the women were living in their parents' home.  Therefore, it seems likely to conclude that at least some of the child-rearing responsibility would be shared with the woman's parents.
Procedures and Attrition
 After randomization clients were given an intake interview and referred to a case manager.  The case manager interviewed the client more extensively to determine a treatment program.  The treatment program for the experimental clients was designed to take maximum advantage of available services and to last for one year.  Control clients received customary care.  Every six months after intake each client was reinterviewed.
 – TABLE 4 ABOUT HERE --
 Table 4 describes attrition.  Extraordinary efforts were made by staff to find and keep in touch with clients.  Attrition only reached about 20% in both groups after two years, a very low figure for this type of experiment.  In part the low attrition was due to the youth of the clientel, many of whom resided with their families or in the same neighborhoods as their families.  The most frequent reason for attrition was departure from the area (6 of 17).  Of the eleven remaining clients 8 refused to cooperate and three could not be found.
Analysis
 Differences between groups was calculated at six month intervals for three years from the beginning of service.  All experimental clients terminated service at the twelve month mark (their average length of care was 11.1 months).  Control clients averaged 2.6 months in treatment.  The findings are presented below as differences between groups at the two year follow-up.
 FINDINGS
 – TABLE 5 ABOUT HERE --
 Table 5 displays the outcomes for experimental clients and control clients at the two year follow-up, that is, at the thirty seventh month after entering care.  At termination of services,  experimental clients were functioning in a number of important areas (% married, social adaptation, depression, and % currently in school) at statistically significant levels above the controls.  As Table 5 reports, these gains were enhanced and new ones emerged relative to the control clients two years after treatment.  Experimental clients were far superior to controls relative to financially independence, public assistance, working and earnings, emotional(HRSD) and social(SAS)adjustment, and happiness(GSS).  Two years after terminating treatment, the experimental group was statistically and substantially better off than the controls in virtually every area of functioning except for living arrangements.  Moreover, both groups appeared to have made gains over their initial statuses although these improvements are ambiguous without true nontreatment controls.
 Yet the most important finding of the study in light of current social concerns was that experimental clients experienced only one half as many out-of-wedlock second pregnancies as control clients.
 These results appear to be related to the services that the clients received.  Of particular interest in light of the current concern over repeated out-of-wedlock pregnancies, it appears that education, family planning, and social work counseling were particularly useful services in reducing the number of single clients who gave birth (or intended to go to term) for a second time.
 DISCUSSION
 In contrast with similar programs (notably New Chance and and Project Redirection but others too)it is apparent that social work services -- counseling, case-management, referral, and follow-up, made a large difference in the quality of life of unmarried teen-age mothers.  In almost every area of concern, participants who received the intensive care of the experimental intervention improved in comparison with the low service intensity controls.  Social work services also achieved the central goal of the program -- decreased incidence of repeat out-of-wedlock births.
 These findings are particularly valuable since the research itself incorporated design features -- notably randomization, multiple objective measures, and relatively large samples -- that lend credibility to its outcomes.  Indeed, these findings are provocative enough to warrant broader attention to the value of making intensive social work more generally available.
 The findings tend to contradict a number of unfavorable tests of intensive social services.  From best estimates of the Agency (total costs divided by number of clients) it seems that the average cost of care for control clients was about $576 and for Experimental group clients about $3,975.  The $3,419 difference seems well justified by the outcomes.
 REFERENCES
 Fraser, M. W., P. J. Pecora, and D. A. Haapala. 1991. Families in Crisis New York: Aldine de Gruyter.
 Friedman, H. 1975. "Why are they Keeping their Babies?" Social Casework. 20(July):322-323.
 Gallagher, U. M. 1973. "Changing Focus on Services to Teenagers," Children Today (September-October).
 Gueron, J. 1990. "Work and Welfare: Lessons on Employment Programs," Journal of Economic Perspectives.
 Haring, B. 1975. "Adoption Trends, 1971-74," Child Welfare 54(July):524-525.
 Maynard, R. 1995. "Teenage Childbearing and Welfare Reform:  Lessons form a Decade of Demonstration and Evaluation Research," Children and Youth Services Review 17:309-332.
 Polit, D. F., J. C. Quint, and J. A. Riccio. 1988. The Challenge of Serving Teenage Mothers: Lessons from Project Redirection New York: Manpower Demonstration Research Corporation.
 Polsby, G. K. 1974. "Unmarried Parenthood: Potential for Growth," Adolescence 9(Summer):273-284.
 Quint, J. C., J. A. Riccio, and D. F. Polit. 1997. New Chance: Final Report on a Comprehensive Program for Young Mothers in Poverty and their Children New York: Manpower Demonstration Research Corporation.
 Rancifer, L. M. and C. Jones. [unpublished] "Constructing Outcome Measures for a Pregnancy Prevention Program: Sociometric Properties."
 Schesinger. 1973. "The Unmarried Mother who Keeps her Child," Child Welfare 52(April):230-233.
 Shlakman, 1966. "Unmarried Parentlhood: An Approach to Social Policy," Social Casework, 47(October):494-501.
 Stahler, G. J. and J. P. DuCette. 1991. "Evaluating Adolescent Pregnancy Programs: Rethinking Our Priorities," Family Planning Perspectives 23(3):129-133.
 Stein, T. and E. Gambrill. 1977. "Facilitating Decision Making in Foster Care," Social Service Review 51(September):502-511.
 Ways and Means Committee, U. S. House of Representatives. 2002. The Green Book Washington, D.C.: U. S. Printing Office.
 Wu, L. L. 1996. "Effects of Family Instability, Income, and Income Instability on the Risk of Premarital Birth," American Sociological Review 61:386-406.Table 1 Selected Characteristics of IC and Control Clients at the time of entering care: Mean (Standard Deviation) or %.

Characteristic     Experimental      Control
        Mean   SD   Mean  SD
          (n=41)      (n=41)

Demographics
 
Age (years)       16.8 (2.4)       16.5 (2.6)
%African-American         41.7      43.2

%Years of schooling completed         9.4 (3.8)      9.7 (3.6)

%Currently in school     38.3          41.2

Living arrangements
 % with parent(s)     62.4      68.3   % with other relative           22.7      19.8
 % independent or other    14.8      11.9
 
 Psychological

Global satisfaction score     2.7 (1.4)      2.5 (1.6)

Hamilton Rating Scale for     15.3 (3.4)       14.7 (3.2)
 Depression

Interpersonal and Social

Contact with father
 Extensive         5.0   7.6
 Minimal        25.6      28.4  Virtually none       69.4      64.0

Father providing support      10.6      12.8
 (more than $20/wk)

Many friends        45.3      42.1

Dating actively       63.5      60.7

Sexual intercourse       45.9          48.4

New pregnancy to term       0.0   0.0

% Married          0.0       0.0

Social adjustment scale       2.2 (.59)  2.3 (.61)

Employment and income

Hours/wk working        5.4 (3.5)      6.5 (4.1)

% receiving public assistance     72.4            76.4
 (AFDC, FS, housing)

Monthly income (previous     387 (40)        382 (47)    3 months)


Table 2:  Receipt of services by experimental and control groups during the year of care by percent of clients in each group accepting service.
 

Service       Control Experimental Significance

Housing      2.1  64.3  ***

Economic/financial    4.2  57.3      ***

Employment        10.9  86.4  ***

Education and GED   74.2  97.2  **

Physical health/medical care 10.9  74.5  ***

Mental health        22.3  41.4  **
 
Homemaker/housekeeping   3.1  50.2  ***

Drug/alcohol               31.2      35.6  ns

Family planning        87.4  97.2   ns

General parenting skills       9.2      62.5  ***

Unspecified assessment       15.9  42.7  ***
  /counseling
 

* less than .05
** less than .01
*** less than .001
Table 3:  Average number of hours of service per client received by experimental and control groups during the year of care.
 
 

Service      Control  Experimental Significance
 

Employment    2.6  26.7       ***

Additional education    47.4     185.3    ***
 and GED prep

Physical health/medical     14.7        53.2    ***

Mental health      11.2        26.3    **
 
Homemaker/housekeeping     5.2        42.2   ***

Drug/alcohol              6.7        18.9   **

Family planning       3.5      3.8      ns

General parenting skills     10.8       33.5   **

Unspecified assessment       6.2       40.7   ***
  /counseling

*   less than .05
**  less than .01
*** less than .001


Pos Table 5.  Selected measures for experimental and control groups 36 months after start-up, by number and significance.
                      Experimental    Control   Sign
 

Outcome
 
Average number of weeks in:

 Any education    73.1   44.2     **
 GED prep     44.3   17.2     *
 Skills training   16.6   11.7

Education attained by month 37 (%):

 GED      62.7   38.1     **
 High school diploma   21.4   10.9     *
 Trade certificate or license 20.5   22.4

Grade level reading by month 37  11.2   7.5     *

Living arrangements by month 37 (%):

 With parent or grandparent     28.6  32.7

 With husband or partner    47.4          24.5     *
 
 With children only             19.8          25.7

Has had trouble locating
 acceptable housing     26.3          61.4      **

Physical and mental health at month 37:

 Health perceived as excellent  64.9  40.2      **

 No insurance or Medicaid    8.2          15.6

 Global satisfaction score       5.6           2.7      *

 Hamilton Rating Scale          25.8          16.9      **
  (for depression)

Social adjustment scale
 at month 37:        39.8         26.7      *
 

Employment and TANF/Welfare receipt:

 Ever employed (%,36 months)      65.7  30.9      **
 

Table 5. Cont'd

 Total earnings (mean, 36 mos.) $27,997        $6,126   ***

 Ever on welfare (%, 36 mos.)      44.5        68.5     **
 
Fertility:

 Ever gave birth (%, 36 mos.)  33.4       61.5     **

 Ever pregnant (%, 36 mos.)  72.7   75.9

 Abortion (%, 36 mos)   65.3       16.2     ***