Pushing Past 'No': The New York Times and Journalistic Ethics

I read this article from New York Times reporter Graham Bowley, which details his attempts to interview Sahar Gul, a fifteen-year old Afghani girl who was tortured by her husband and in-laws after refusing to go into prostitution. I was interested in The New York Times coverage of this horrifying incident, but was shocked with Bowley's reporting. What follows is the subsequent letter I wrote to the editor:

To the Editor:
I was dismayed to read Graham Bowley's (“In One Girl’s Story, a Test of Women’s Rights in Afghanistan”, January 10, 2012) article describing his interview with Sahar Gul, the Afghan teenager who was the victim of gruesome abuse by her husband and in-laws. I believe the manner in which Mr. Bowley attempted to gain access to Ms. Gul while she was recovering in the hospital is unprofessional and unethical. Yet, The New York Times felt it was appropriate to publish these details. He took great pride describing how he “pushed past ‘no’” when hospital workers repeatedly told him that Ms. Gul did not wish to speak to reporters because she was too emotionally fragile to revisit her story. Mr. Bowley notes that Ms. Gul had already been interviewed by another news organization (the Associated Press), and therefore, the goal of disseminating her story with a broader audience—indeed, an important goal for journalism—was already achieved. However, Mr. Bowley’s persistence in interviewing Ms. Gul seemed to be more of an attempt to advance his professional career rather than to share an important story with the world. I see no need for a The New York Times reporter to subject a child to a additional questions and potentially retraumatize her in the process just to enhance a newspaper’s credibility. There is nothing to be gained in this process, and journalists such as Mr. Bowley must remember that they have an ethical responsibility when reporting in volatile settings such as Afghanistan.
Regards,
Bree Akesson

I shortly received the following response from The New York Times public editor, Arthur S. Brisbane, who "works outside of the reporting and editing structure of the newspaper and receives and answers questions or comments from readers and the public, principally about articles published in the paper":

Thanks for your message about Graham Bowley's coverage of Sahar Gul, the young Afghan girl. I am concerned about the girl's privacy as well and have raised the question with the Foreign Desk. I do concur that news organizations should be careful to respect the privacy of crime victims. This is a case where, I believe, the benefits of doing a story were outweighed by the potential harm to the girl.

Best,
Art Brisbane public editor

Arthur Brisbane wrote an-ed piece about this topic, addressing many of the concerns in my letter (which reflected many other readers' views), along with a response from the foreign editor. You can view the article here.

Study Finds Sexually Abused Children in Turkey Exhibit More Psychiatric Disorders at Initial Assessment than Two-Years Later

Prior research indicates that children who have been sexually abused are at risk of developing various negative short and long-term psychological sequelae. However, there is little research on the effects of sexual abuse on children living in non-Western contexts. In this study, Ozbaran et al. (2009) aim to explore the emotional and psychological impact of sexual abuse among a sample of children in Turkey at referral and two years after referral. Over 300 parents and their children (ages 5-16) were referred to the Department of Child and Adolescent Psychiatry at Ege University in Turkey, and 20 parents and their children agreed to participate in the research. At referral, the parent-child pairs (N=20) completed the first evaluation, which included a detailed psychiatric examination of each participant, completion of the Child Behavior Checklist by the mother, and an evaluation of the child’s mental capacity by a clinical psychologist. Over a period of two years, the sample participated in a combination of the following interventions, depending on the child and family’s needs: psychopharmacologic medication, parental support groups, family meetings with social workers, school counseling services, sport and art activities, and outpatient occupational therapy. After two years, the parent and child were asked to return for a follow-up evaluation. The study found that children who were sexually abused had more psychiatric disorders during the first assessment (55%) than during the second-stage evaluation (0%). However, the study also found that there was no statistically significant relationships between psychiatric diagnoses, socio-demographic features, duration of abuse, or abuse type.

The study methodology lacked rigor in that it was not clear that there was a systematized method for conducting the first and second-stage evaluations. However, the implementation of blind psychiatric interview during the second-stage evaluation was strength of the study, because it lessens the likelihood that the therapist’s impressions will be biased. The small sample size (N=20) of the study, ensured that no participants were lost to follow-up. However, the small sample size - and therefore the decreased statistical power - also made it difficult to draw any statistically significant conclusions. The lack of a control group in the study design also made it a challenge to directly demonstrate a causal relationship between sexual abuse and development of a psychiatric diagnosis or behavioral problems. The authors’ statement that the condition of the sexually abused children in their sample improved significantly as a result of time is misleading, because the study design doesn’t discern whether this is attributed to time or to the array of intervention services that the children and their families received during the study period. Lastly, sampling bias may pose a threat to the external validity of the study, as the sample reflected less than 7% of the initial referrals (over 300) to the hospital. This small sample may have characteristics that the general population does not exhibit. In light of these limitations, this study still adds to the sparse literature on the topic of children affected by sexual abuse in non-Western settings and represents an obvious precursor to more rigorous studies.

Source:
Ozbaran, B., Erermis, S., Bukusoglu, N., Bildik, T., Tamar, M., Ercan, E.S., Aydin, C., and Cetin, S.K. (2009). Social and emotional outcomes of child sexual abuse: a clinical sample in Turkey. Journal of Interpersonal Violence, 24(9), 1478-1493.

Mind-Body Skills Intervention Decreases Symptoms of Post-Traumatic Stress Disorder in Postwar Kosovar Adolescents

Incidence of post-traumatic stress disorder (PTSD) has been reported among war-affected children and adolescents. Yet few studies have evaluated the effectiveness of PTSD treatment programs for this population. Gordon, et al. (2009) conducted a randomized control trial to determine the effectiveness of an intervention aimed at ameliorating the effects of PTSD in adolescents. All children in Jeta e Re (“New Life”) High School in Kosovo were screened to participate in the study. Eighty-two students met the criteria for PTSD and were randomly assigned to two groups: (a) intervention group and (b) delayed intervention group. The intervention group participated in a comprehensive mind-body skills group program, a 12-session program consisting of meditation, guided imagery, breathing exercises, relaxation activities, therapeutic discussion, and art therapy. The delayed intervention group received the same intervention once the intervention group had completed the program, approximately 6 weeks later. The study shows that students receiving the mind-body skills group program had significantly reduced levels of PTSD symptoms as compared to the delayed intervention control group.

This study is the first randomized control study to examine the effects of a therapeutic intervention on war-affected adolescents. The study design overcomes the ethical dilemma of withholding promising treatment from the control group by providing treatment to the control group after the intervention group has completed post-intervention interviews. Nevertheless, expectation bias poses a threat to the study’s internal validity, because the results of the improved outcomes may be due to the participants’ expectations that the intervention would be effective. This threat to internal validity is supported by the fact that the intervention was piloted in the school prior to the study, building even more anticipation, as students are bound to share their experiences with other students. This is also be related to the process of testing, which could be a threat to internal validity, because students may become familiar with the measures being used. Because teachers both facilitated the intervention and administered the measures, unintentional expectancy effects pose a threat to the study’s construct validity. In this case, teachers may have subconsciously expected students to show fewer PTSD symptoms in the post-test. Similarly, the students may have wanted to please the facilitators by showing improvements in the post-test. Despite the authors’ explanation that these students would only be comfortable speaking with a familiar adult, future studies should utilize external experimenters to administer the pre- and post-test, because they will be naïve to which students received the intervention and will have no vested interest in the outcome of the intervention.


Reference:
Gordon, J.S., Staples, J.K., Blyta, A., Bytyqi, M., and Wilson, A.T. (2008). Treatment of posttraumatic stress disorder in postwar Kosovar adolescents using mind-body skills groups: A randomized controlled trial. Journal of Clinical Psychiatry, 69(9), 1469-1476.

Study in Afghanistan Finds Trauma Exposure and Caregiver Mental Health Predict Risk of Negative Mental Health Outcomes in War-Affected Youth

Research conducted in war-affected situations has found an association between exposure to traumatic events and mental health outcomes. However, there have been few studies focusing on the mental health outcomes of children and adolescents affected by war. Panter-Brick, et al. (2009) conducted a study aimed at determining a causal relationship between war and mental health problems, specifically looking at associations related to gender, traumatic events, caregiver mental health, and socio-demographic characteristics. The sample targeted children ages 11 to 16, who were randomly drawn from randomly selected schools in three purposively chosen regions of Afghanistan. Data were gathered from three informants (children, caregivers, and teachers), who were interviewed with a variety of brief measures. Measures were chosen because they had high reliability in other contexts of instability. To increase instrument diagnostic validity, measures were translated, back-translated, and reviewed by a multi-disciplinary team. The research found that the variables of exposure to multiple trauma and caregiver mental health were predictive of psychopathology for in-school Afghan adolescents. Previous research corroborates this finding and adds to the study’s external validity. Furthermore, the study found that girls were more likely than boys to develop poor mental health outcomes, such as depression.

The authors explore the quality of traumatic events to include not just war-related events, but accidents, corporal punishment, illness, death of a relative. The quality of the particular incident is not indicative of predicting poor mental health, but rather the accumulation of traumatic events contribute to risk factors for mental health problems. The study includes a few threats to validity. By only sampling school children, the study neglects out-of-school youth, who may be at high risk of developing psychopathology. Furthermore, even though the sample of in-school youth was stratified to include accurate proportions of boys and girls, the focus on in-school youth disproportionately excludes girls, because boys are more likely to attend school than girls. The school samples were restricted to three urban areas, affecting external validity and indicating that future research might consider including a sample from a rural setting. Despite its limitations, this study indicates the value of working within the school system to assess children’s mental health needs and proves the value of school-based mental health services. Furthermore, the findings contribute to the small but much needed knowledge base about the mental health of war-affected children.

Reference:
Panter-Brick, C., Eggerman, M., Gonzalez, V., and Safdar, S. (2009). Violence, suffering, and mental health in Afghanistan: a school-based survey. The Lancet, 374, 807-816.

Preventive Interventions Help Secure Attachment in Maltreated Infants

Child maltreatment presents threats to child developmental processes, including long-term maladjustment and development of psychopathology. Cicchetti, Rogosch, and Toth’s (2006) study examines the effectiveness of two preventive interventions: psychosocial parenting education (PPI) and infant-parent psychotherapy (IPP). Based on the theory that secure attachment is derived from parental knowledge and skills, PPI involves a series of psychoeducation-focused home visits with mothers to improve parenting skills, increase mothers’ knowledge of child development, and enhance coping and social support skills. IPP is based on attachment theory and includes an exploration of the parent-child relationship through therapy. The hypothesis that PPI or IPP interventions will be most effective for ameliorating the effects of maltreatment of infants is explored in this study. The study sampled mothers and their infants in maltreating families (n=137). Subjects were randomly assigned to one of three cohorts: (1) psychosocial parenting education (PPI), (2) infant-parent psychotherapy (IPP), or (3) community standard (CS). A normative comparison (NC) group of non-maltreated infants and their mothers (n=52), with similar demographics to the intervention group, was also included. Using validity tested measures, the authors found that mothers and children in the PPI and IPP groups showed greater increases in secure attachment than mothers and children in the CS and control groups. This study supports evidence that preventive intervention programs can alter attachment organization and ameliorate the negative developmental consequences of maltreatment.

As compared to results from the CS group, the authors provide evidence that the PPI and IPP interventions result in greater increases of secure attachment in maltreating families. However, the non-maltreating control group varied greatly from the maltreating group, with maltreating mothers reporting higher rates of maltreatment in their own childhoods, more insecure relationships with their mothers, more maladaptive parenting attitudes, more parenting stress, and lower family support. Similarly, despite trying to maintain uniform adherence to the therapeutic intervention, the therapy was not completely standardized, because the intervention used different therapists. Lastly, the intervention groups were actively pursued to follow the therapy schedule, which is not commonly done in treatment modalities such as the CS group.
The authors attempt to eliminate any self-selection bias, by seeking out maltreating families rather than the families seeking to be enrolled in the study.

Reference:
Cicchetti, D., Rogosch, F.A., and Toth, S.L. (2006). Fostering secure attachment in infants in maltreating families through preventive interventions. Development and Psychopathology, 18, 623-649.