A 30-Month Follow-Up of Court- Referred Batterers in Four Cities

• Date: 01 February 2000
• Published date: 01 February 2000
• DOI: 10.1177/0306624X00441010
• Author: Edward W. Gondolf
• Subject Matter: Journal Article

InternationalJournalofOffenderTherapy and Comparative Criminology

Batterer30-MonthFollow-Up

A 30-Month Follow-Up of Court-

Referred Batterers in Four Cities

Edward W. Gondolf

Abstract: A follow-up of court-referred male batterersin four cites was conducted to assess

thelong-term outcomesof battererprograms (N=618). About70% ofthe batterers’initialand

new female partners were interviewed by phone every 3 months for the first 15 months and

every 7 to 8 months for the next 15 months. Accordingto the women’s reports, 41% of the men

committed a reassaultduring the 30-month follow-up period—an increase of only 7% over the

15-month reassault rate. Nearly two thirds of the first-timereassaults occurred in the first 6

months. About a fifth of the men repeatedlyreassaulted their partners and account for most of

the reported injuries. The reassaultrate for men attending a program for 3 months or more is

significantly less than for dropouts. The reassaultrates are not significantly different across

the four programs, despite differencesin program length and services.

Theextendedoutcome ofbattererprograms has beena topic ofdebateamong bat-

tered women’s advocates and batterer program participants (Hamberger& Hast-

ings, 1993). On one hand, many advocates arguethat batterer programs have little

impact on participants beyond the conventional follow-up of 6 months to a year.

Batterersmaycontain theirbehaviorwhile underprogramsupervision orthreatof

sanctions, the advocates contend, but are likely to return to a pattern of abuseafter

ayearor so. Inother words,reassaultrates escalate substantiallyover time.Onthe

other hand, many who work directly with batterers believe that a majority of the

men do not reassault their partners, and a portion of men make substantial behav-

ioral changes. Education from the program, sanctions from the court, and circum-

stancesofthe relationshipcombineto contain,interrupt,or stopabusefor thelong

term. That is, reassault rates continue to decrease beyond a year or more after pro-

gram intake.

These competing clinical observations conform to theoretical conceptions of

criminal interventions with other populations. Deterrence theory suggests that

abuse may escalate following an abbreviated follow-up (i.e., 6-12 months) of

intervention. The “Scared Straight” programs with juvenile delinquents are

NOTE:This researchwas madepossible throughagrant fromthe Centersfor DiseaseControl andPre-

vention(CDC), U.S.Department ofHealth andHumanServices (GrantNo. R49/CCR310525-02),but

itdoes notnecessarilyrepresent theofficialview ofthe CDC.Previous versionsof thisreport werepre-

sented at the Program Evaluation and Family ViolenceResearch Conference, Durham, New Hamp-

shire,July26-29, 1998;and theannual Meetingof theAmerican Societyof Criminology,Washington,

D.C., November 11-14, 1998.

International Journal of Offender Therapy and ComparativeCriminology, 44(1), 2000 111-128

2000 Sage Publications, Inc.

111

probably the most popularized example of this outcome (Finckenauer & Gavin,

1998). Behavioral therapy models suggest that men gradually learn to interrupt

problematic behavior and gradually contain or avoid that behavior. For instance,

an extended follow-up of patients in a 1-month intensive residential treatment

program for alcoholism showed a sharp decrease in alcohol useover time. More

than two thirds (71%) of the patients were using alcohol at 6 months after treat-

ment, 54% at 12 months, and 47% at the extended follow-up (i.e., an averageof 9

years) (Shaw, Waller, Latham, Dunn, & Thomson, 1997).

Extended longitudinal follow-ups are fairly routine in injury, disease, and

criminological research, but the research on woman battering has been confined

largely to 12- to 18-month follow-ups (Rosenfeld, 1992; Tolman & Bennett,

1990). Some follow-ups of batterer programs have been longer, but these have

beendonewith smalllocalsamples anda range offollow-up periods(e.g.,Dutton,

1987; Dutton, Bodnarchuk, Kropp, Hart, & Ogloff, 1997; Saunders, 1996;

Shepard, 1992). The extended follow-ups also rely on recidivism identified in

criminal records rather than by victim reports. The longest extended follow-up

was conducted by Dutton et al. (1997) with a total of 446 batterer program com-

pleters and dropouts in Canada. The average length of the follow-up period was

5.2years,but thefollow-upwas aslong as 11years for somesubjects. A quarterof

the sample was arrested for at least one assault, and 18% specifically for wife

assault, during the follow-up. There was no significant difference in the recidi-

vism rates for completers and dropouts, but the completers had fewer total re-

arrests for assaults. As the researchers themselves observed, the recidivism rates

grossly underestimate the victim-reported reassaults. Ultimately, an extended

follow-up using victim reports, a uniform follow-up period, and multiple sites is

needed to better assess the extended outcome of batterer programs.

We previously conducted a longitudinal 15-month follow-up of four batterer

programs, using victim reports and arrest records (N= 840), with a response rate

of 69% for the full 15 months). The findings suggest a marked de-escalation of

new reassaults (Gondolf, 1997) and similar outcomes regardless of program

length (Gondolf, 1999). An extended follow-up was proposed to determine

whether the 15-month trends would continue in the long term. An escalation

of reassault might appear after program participation and probation ended, and

the effects of the longer programs (5.5 months and 9 months versus 3 months)

might take longer to appear. In their methodological reviewof domestic violence

arrest evaluations, Garner,Fagan, and Maxwell (1995) argue that the prospect of

reassault escalation needs to be considered to meaningfully evaluate batterer

intervention.

Thisreportpresents a30-monthfollow-up (i.e.,2½years afterprogram intake)

of the court-referred batterers in our original 15-month evaluation. It summarizes

the trend of reassault over the extended follow-up for the full sample and com-

pares the cumulative reassault rates for the four programs. The extended

follow-up also examined other forms of abuse reported by the female partners of

112 International Journal of Offender Therapy and Comparative Criminology