As of October 9, 2009, 463,000 children were receiving care in the public foster care system (U.S. Department of Health and Human Services, 2009). Parental rights to 75,000 (16%) had been terminated. Parents with serious mental illness (SMI) are confronted with increased risk of involvement with the child welfare system and of having their children placed in out-of-home care (Park, Solomon, & Mandell, 2006; Sands, Koppelman, & Solomon, 2004). In fact, five states (Alaska, Arizona, California, Kentucky, and North Dakota) and Puerto Rico list mental illness or disability as grounds for not providing reasonable efforts toward reunification (Kaplan, Kottsieper, Scott, Salzer, & Solomon, 2009). The reality of threats to the parenting status of people with mental illness, particularly the influence of stigma on policy and services decisions, makes this an important area for social work research. Because much of the research on custody loss among women with SMI is dependent on self-reports, valid measures are imperative. In the present study, we were interested in discovering whether a Life History Calendar (LHC) would provide a valid measure of custody loss. Using a sample from a longitudinal study of women with SMI who were parents, we compared their responses on an LHC with their responses to a standard structured interview questionnaire. METHOD Sampling The Birth Mothers. The longitudinal study in which participants were involved was a study (RO1-54321) of the parenting history of women with SMI. Serious mental illness was defined as having been diagnosed with schizophrenia, schizoaffective disorder, bipolar disorder (with or without psychosis), or major depression (with or without psychosis). Participants were recruited from 12 community mental health centers and from the inpatient psychiatric units of three hospitals, all in southeast Michigan. Eligibility criteria were as follows: being 18 to 55 years of age, having primary care responsibility for at least one child four to 16 years of age, having a severe mental illness lasting at least one year and producing major dysfunction in one or more life areas, and being in active mental health treatment. Of 379 participating at baseline, 324 were interviewed at wave 2 and 322 at wave 3. Wave 2 and wave 3 participants did not differ significantly demographically from those at wave 1. (Demographic data are provided for wave 3 participants.) All participants provided voluntary informed consent according to the criteria of our university human subjects review board. Wave 3 participants had a mean age of 40 years and 3.24 children on average. Sixty percent were black/African American; the next largest racial group (30%) was white/Caucasian. Among birth mothers still participating at wave 3, 36% had less than a 12thgrade education, 23% had completed high school, and 41% had completed education beyond high school. Diagnosis was measured at wave 1 only, at which time information from 28 interviews was not sufficient to reach a diagnosis. Of the remaining, 10% of the wave 3 participants had had a wave 1 diagnosis of schizophrenia/schizophreniform disorder, 11% of schizoaffective disorder, 40% of major depression, 12% of major depression with psychosis, 12% with bipolar 1 disorder, and 15% with bipolar 1 disorder with psychosis. Seventy-four percent considered themselves still receiving mental health services at wave 3, and 70% said they were being prescribed medication for their psychiatric disorder. The Children. In the study, 575 children were represented. (This excludes children who were not the mothers' birth children, were older than 18, or, in one instance, for which no custody loss information was available.) Of the 575, 56% were boys; the mean age was 10.6 years. At wave 2, birth mothers reported currently having custody of 84% of their children. Procedure Data Collection. Interviews were conducted in participants' homes. …