ABSTRACT
Can standardized assessment contribute to improving Adult Protective Services (APS) practice? In this exploratory study, San Francisco and Napa APS utilized a newly developed short self-neglect assessment to test how standardized measures provide information for substantiation decision making. Findings demonstrated satisfactory reliability and validity for the short self-neglect assessment, and analyses revealed important issues that could improve practice. Review of outliers revealed: (1) problems using the assessment tool, (2) misunderstandings of APS procedures, and (3) struggles navigating the case management system. These revelations could all be easily addressed in training. Caseworker’s clinical judgment and experience also continue to have a place in complementing the use of the standardized assessment.
The need to move toward a systematic decision-making system is widely acknowledged as the desired standard practice in adult protective services (APS) (Connolly, Brandl, & Breckman, Citation2014). In fact, case decisions based on clinical judgment alone have been found to have little predictive validity for future harm (Park, Johnson, Flasch, & Bogie, Citation2010). Additionally, unreliable and/or inconsistent case assessments may result in very different intervention recommendations based on the worker assigned to the case (Mosqueda et al., Citation2016). As a result of such findings, the National Council on Crime and Delinquency launched projects that introduced “structured decision making” (SDM) into APS practice. The desired outcome of SDM is to increase consistency and accuracy in assessing self-neglect and elder mistreatment to improve standardization (Park et al., Citation2010). However, to our knowledge, findings of beneficial outcomes such as the elimination and reduction of abuse, even from these latest structured approaches, have not been published.
Clinical judgment should be highly regarded in APS practice. For example, some victims may lie out of fear or concern for their family who are abusing them. In such cases, the caseworker must see beyond the false and defensive statements of alleged victims and abusers alike. In unclear situations like this, having standardized measures provides scaffolding in the form of common items for case investigation and decision making, so that caseworkers are not left without guidance about key areas to discuss and to make judgments. Therefore, advantages of standardization include: (1) improved quality control by “speaking the same language” among caseworkers and supervisors; (2) use of data to provide feedback about how caseworkers are using the system, and to detect errors and problems for improved decision making; (3) identification of abuse types and severity through quantifiable measures, and subsequently, to reassess using the same measures to check for improved or worsened outcomes; (4) implementation of best practice across APS agencies (Conrad, Iris, & Liu, Citation2017). Although many barriers still exist that prevent valid outcomes to be assessed, promoting accountability and feedback for improvement has been advocated (e.g., Conrad et al., Citation2017) through ongoing projects such as the use of Identification, Services, and Outcomes (ISO) Matrix in San Francisco and Napa Counties in California (Liu, Conrad, Conrad, & Hass, Citationin progress).
The ISO Matrix is a decision-support system designed for APS to capture outcomes. It has three parts: (1) identification of abuse, which includes the Elder Abuse Decision Support System (EADSS) Short Form (Beach et al., Citation2017), client and abuser risk assessments (Conrad & Conrad, Citation2019), and new assessment tools investigating self-neglect, isolation, abandonment, and abduction during case investigation; (2) services that are documented by caseworkers to address the types of abuse identified; (3) outcomes that are estimated by the change score from the pretest identification of abuse to the posttest at case closure, using the same measures as at pretest during case investigation. Since the ISO Matrix’s assessment tools are mostly based on the EADSS, which is a theory-based system developed through extensive literature review, concept mapping, and testing in the field (Conrad et al., Citation2017; Conrad, Iris, Ridings, Langley, & Anetzberger, Citation2010; Conrad, Iris, Ridings, Langley, & Wilber, Citation2010; Conrad, Liu, & Iris, Citation2019), the ISO Matrix inherits the EADSS’ comprehensive and structured interview guides to assess the harm of abuse and client/abuser characteristics. Additionally, the ISO Matrix expands the EADSS by adding additional assessment tools to investigate other types of abuse in the field. The pretest and posttest design of the ISO Matrix also opens the possibility for APS to interpret the effects of services.
This report tests the hypothesis that standardized measures can be used in APS to examine caseworker practice in substantiation as well as to provide feedback about system functioning that can be useful for improved practice. In this study, the ISO Matrix assessment tools were used for the investigation to substantiate elder abuse and to estimate the type and severity of the abuse. Using the scores, it was possible to examine the distribution of scores in each of the three categories of findings: unfounded, inconclusive, and confirmed, to see if this process could help the caseworkers make better decisions. Specifically, using self-neglect as an example, substantiation decisions were plotted against quantitative measures using the ISO Matrix to examine whether the information could be used to improve substantiation decision-making. Since a short-form measure for self-neglect that is feasible to be used by APS had not been developed nor validated, the reliability and validity of the self-neglect short form in the ISO Matrix were analyzed.
Methods
Participants
San Francisco and Napa APS caseworkers (43 from San Francisco, 7 from Napa) and supervisors (9 from San Francisco, 1 from Napa) participated in a six-month pilot demonstration to use the ISO Matrix from August 2018 to January 2019. A total of 2,063 face-to-face cases had the ISO Matrix collected during this period of time (1,819 from San Francisco, 244 from Napa).
Procedures
Caseworkers and supervisors attended a day-long training on APS outcomes and the ISO Matrix, including how to use the ISO Matrix in the field and document findings in their respective county’s case management system. It was emphasized that the standardized measures of the ISO Matrix aim to improve practice consistency instead of substituting for caseworkers’ clinical judgment. Therefore, even though ISO Matrix items are framed in a way that could be asked of alleged victims directly, all answers on the ISO Matrix were to be based on caseworkers’ investigation rather than solely on alleged victims’ comments.
This study’s focus is on the scores collected during case investigation and their relationships to caseworkers’ substantiation decisions. The Institutional Review Board of Purdue University (IRB Protocol # 1812021397) deferred the approval to the University of California, San Francisco (IRB # 17–23904) to provide annual oversight.
Measures
Self-neglect short-form measure
The neglect section of the EADSS Short Form was modified for self-neglect case investigation including six questions (see Appendix A). Two items were deleted from the EADSS Short Form since they described a caregiver neglect situation, which is not suitable for self-neglect cases. An additional consideration for the short form was that previous work on self-neglect has shown that there are two components to the construct – behavioral and environmental (Iris, Conrad, & Ridings, Citation2014). As indicated in Appendix A, yes was scored as 2, some indication was scored as 1, and no was scored as 0. Caseworkers also had the choice for don’t know (did not get this information) and refused to be answered by the client, both of which were coded as missing data. The six items were summed for the ISO Matrix self-neglect score ranging from zero to 12.
Self-neglect substantiation
As mentioned, three categories of substantiation decision are used in California: unfounded, inconclusive, and confirmed. During the pilot demonstration, five subtypes of self-neglect were collected, including physical care, medical care, health and safety hazards, malnutrition/dehydration, and financial. Each subtype has a substantiation decision if an allegation was reported. To simplify the data analysis for this study, as long as one subtype was confirmed for an individual, the case was considered to be confirmed of self-neglect. If no subtypes were confirmed and at least one was inconclusive, the case was considered inconclusive of self-neglect. When all self-neglect subtype allegations were unfounded, the case received an unfounded substantiation. Out of the 2,063 face-to-face cases, 1,192 cases had a self-neglect substantiation decision. The other 871 cases were not alleged with self-neglect, and investigation of other alleged types of abuse did not find any indication of self-neglect. In other words, 58% of cases were alleged or found to have at least one of the self-neglect subtypes (see Appendix B for distribution by score and substantiation decision).
Statistical analysis
Reliability and validity of self-neglect short form measure
Following Beach et al. (Citation2017), the predictive validity, as measured by area under the Receiver Operator Characteristic (ROC) Curve, and internal consistency, as measured by Cronbach’s alpha, of the self-neglect short form were evaluated. The area under the ROC curve (AUC) can be estimated using a logistic regression model with the substantiation decision as the dependent variable (positive: confirmed and inconclusive/negative: unfounded) and the total score from the six self-neglect items as the independent variable. Hosmer and Lemeshow (Citation2000) noted that an AUC of 0.5 suggests no discrimination (i.e., no ability to diagnose patients with and without the disease or condition based on the test); 0.7 to 0.8 is considered acceptable; 0.8 to 0.9 is considered excellent; and more than 0.9 is considered outstanding.
Box and whisker plots
Box and whisker plots are descriptive statistics used to detect sample distribution. In the box and whisker plots charted in , we see the three categories along the horizontal or x-axis: unfounded, inconclusive, and confirmed. Along the vertical or y-axis, we see the scores running from zero to 12 on the self-neglect measure of the ISO Matrix. For each box and whisker plot, the top of the dotted whisker represents the 75th percentile, the dark line in the middle is the median (50th percentile), and the bottom of the dotted whisker is the 25th percentile. Individual circles represent outliers, and they were investigated through case reviews and interviews with caseworkers.
Figure 1. Box and whisker plot by ISO Matrix score and substantiation category.
Case review and caseworker interview
To examine outliers and other unexpected findings in greater detail, we randomly selected cases with ISO Matrix self-neglect score outliers for a case review. This was done to garner caseworkers’ feedback on the ISO Matrix and to better understand their decision-making process for substantiation decisions. Caseworkers with more than one outlier cases were invited to discuss their substantiation decision-making process in light of the ISO Matrix self-neglect scores. However, the invitation did not inform caseworkers that they were invited because of multiple outliers within the same substantiation category to avoid defensive attitudes toward being interviewed. Participation was voluntary and confidential. Some caseworkers declined to participate due to heavy workload or leave schedule.
Results
Validation of self-neglect short form measure in ISO Matrix
Predictive validity of the scores on the six self-neglect items was tested against caseworkers’ final substantiation decision of self-neglect using area under the ROC curve. There were 1,013 positive self-neglect substantiations and 179 negative. The area under the ROC curve was high at 80%, indicating that the six-item self-neglect short form did a good job of identifying self-neglect cases. The Cronbach’s alpha for the six-item measure was.74, which is acceptable and similar to the neglect short form that had a Cronbach’s alpha of .77 from Beach et al. (Citation2017). All of these data provided evidence of the validity and reliability of the six-item self-neglect short form.
Findings of box and whisker plots
In , the left “box” (a single horizontal line), which is the unfounded substantiation decision, had a majority of cases at zero. The middle box, which is the inconclusive substantiation decision, had a median of one and approximately 25% of cases with a score higher than two. The middle box experienced a floor effect, i.e., for the inconclusive substantiation, about 50% of cases had scores of zero or one. The right box, indicating the confirmed substantiation decision, had a median of three and had greater variability in scores. Twenty-five percent of cases had a score higher than six, and the bottom 25% had a score of zero or one.
illustrates that each category had distinct score distributions for the measure relative to the other categories. Looking first at the inconclusive category in the middle, the median coincided with the 25th percentile of the confirmed category and is higher than the median of the unfounded category. Put simply, this would inform caseworkers that, in the inconclusive category, cases scoring above the median, were more like confirmed cases, and those scoring below the median are more like unfounded cases.
Reviews of unexpected findings in each category
Review of unfounded cases
Unfounded cases primarily had a zero for ISO Matrix scores in the left box and whisker plot; however, 10 cases had an ISO matrix score higher than three with unfounded findings. Since confirmed cases had a mean of three on the ISO Matrix scores, case reviews were completed on these 10 cases.
Basing ISO Matrix scores on allegations before finishing investigation
In 3 of the 10 cases, the caseworker filled out the ISO matrix before completing the investigation. This means the ISO Matrix scores were based on the uninvestigated allegations in the original report of abuse rather than on the evidence gathered during the investigation. Basing the ISO Matrix scores on the allegations resulted in higher scores than should have been recorded.
Depending solely on client’s report for ISO Matrix scores
In two other cases, the caseworker filled out the ISO Matrix based solely on the alleged victim’s responses. The alleged self-neglecting older or vulnerable adults reported indicators of self-neglect, even when the caseworkers did not find the alleged victims’ statements credible. The higher ISO Matrix scores reflected the alleged victims’ statements rather than the caseworkers’ clinical judgment based on the full investigation.
Marking both self-neglect and neglect when only one was involved
Two caseworkers marked both self-neglect and neglect indicators when they should have only marked neglect. The reason they attributed to the marking of self-neglect was the similar wording on items of these two types of abuse.
Review of inconclusive cases
Two types of cases were examined in the inconclusive categories: high and zero ISO Matrix scores. Looking at the inconclusive column in the center of the box and whisker plots, the outliers at the top are notable. Four cases had high self-neglect scores represented by circles above the whisker in the inconclusive category. These were examined to uncover why they were not classified as confirmed, since these inconclusive cases would be high scorers even in the confirmed category. Likewise, 20 cases with scores of zero were selected at random for case reviews and interviews to discern why they were not substantiated as unfounded.
Reluctance to substantiate as confirmed when clients were not interviewed
In three of the high ISO Matrix score cases, the caseworkers made telephone calls to collateral sources, reviewed records, and gathered evidence. However, the clients either denied or refused to answer the caseworkers’ questions about the allegations of self-neglect and neglect. Even though the clients refused to participate in the APS investigation, it appears from the case record that there was sufficient evidence to confirm the self-neglect allegation. This illustrates where APS caseworkers are, at times, reluctant to make a confirmed substantiation unless the client agrees that the allegations are true and cooperates with the investigative process.
Confusion in using the case management system
The fourth high ISO Matrix score case reviewed involved an elderly client who was referred to APS for possible financial self-neglect. After interviewing the caseworkers, we learned that additional self-neglect sub-types were discovered during the investigation. However, the new sub-types of self-neglect were not documented in the case management system, due to the complex process in adding findings that are not part of the allegation. The case finding was then incorrectly kept as inconclusive based only on the financial self-neglect substantiation while not including the additional sub-types of self-neglect. Including the additional findings would have resulted in a confirmed substantiation decision.
Confusion due to ISO Matrix as a new assessment tool used
The other group of cases that warrant in-depth review under the inconclusive category are those with ISO matrix scores of zero and inconclusive findings, i.e., bottom of the middle box and whisker plot. Four APS caseworkers responded to the request to be interviewed to ascertain their decision-making rationales. The first case involved allegations for two clients, a stepmother for self-neglect and a developmentally delayed stepdaughter for neglect by her stepmother. When the caseworker reviewed this case during the interview, it was indicated that self-neglect substantiation should be confirmed for the stepmother with the appropriate ISO Matrix score instead of zero, and neglect and self-neglect would both be confirmed rather than inconclusive for the stepdaughter with the appropriate ISO Matrix scores instead of zero. The confusing case of both the stepmother and the stepdaughter as clients, while ISO Matrix was newly launched at the time, contributed to the error.
Misinterpreting what to document under “other”
The other interviews with cases of self-neglect that had zero ISO Matrix scores and inconclusive findings indicated that the last item of self-neglect “other” was interpreted by the caseworkers to be limited to the examples listed in the item. Caseworkers suggested that the examples for self-neglect “other” be eliminated to avoid misinterpretation.
Review of confirmed cases
Case reviews were completed on 25 randomly chosen cases identified with ISO Matrix scores of zero for self-neglect but that had confirmed substantiation. Four of the caseworkers agreed to be interviewed to clarify why these cases had zero or no score on self-neglect.
Depending solely on client’s report for ISO Matrix scores
For the first interview, the case had a confirmed substantiation for medical care self-neglect. Like the cases with high ISO Matrix scores but an unfounded substantiation, the caseworker was uncertain on how to complete the ISO Matrix because the client had said the self-neglect issues were resolved before the caseworker made the in-person contact. Even though the issues that caused the referral to APS had been mostly resolved prior to the caseworker’s face-to-face contact with the client, some issues remained that required a confirmed substantiation decision, and during the interview, the caseworker was able to articulate how the self-neglect items should have been marked, such that the score would accurately reflect the confirmed substantiation decision.
Misinterpreting what to document under “other”
The second case interviewed had a confirmed substantiation, but just like the inconclusive cases with a zero ISO Matrix score, the caseworker did not mark “other” in the self-neglect section because “other” was thought to be limited only to the examples listed in the item.
Identifying limitations of ISO Matrix short forms
The other interviews with caseworkers revealed the tradeoff of using the short form because the reason for not having a higher ISO Matrix score was that the caseworkers were unable to find an indicator in the ISO Matrix self-neglect short form that would fit the type of self-neglect identified by the caseworkers. The reasons for not marking self-neglect items in these cases reviewed primarily appeared to be related to caseworkers having a narrower approach to interpreting the wording of the questions in this category. For example, one of the caseworkers confirmed medical care self-neglect in a case, but no self-neglect items were answered as “yes” or “some indication” on the ISO Matrix. The caseworker explained that none of the questions in the self-neglect section are specifically related to medical care self-neglect. The second question of the self-neglect category was interpreted as limited to medication issues or medical equipment but not to medical care. It was noted that the “other” item could have been used and the type of self-neglect described.
Documenting self-neglect findings in case notes instead of ISO Matrix
In four other cases with confirmed self-neglect substantiation decision and no items marked in the self-neglect category, the caseworkers documented their findings in the text box for additional case information. For example, one caseworker had entered “risk of eviction due to nonpayment of rent” and behavior issues in the text box. Therefore, instead of using the “other” category, the explanations were reported only in non-designated areas when they should have been in both.
Discussion
APS caseworkers’ decision of substantiation is based on clinical judgment and evidence found during the investigation. However, without structured approaches, the criteria of each substantiation may differ depending on the caseworker’s experiences. We proposed to use standardized assessment tools to guide caseworkers’ decision-making processes. Ideally, indications from the ISO Matrix would correspond to a positive substantiation decision. Study findings from the ISO Matrix demonstrated satisfactory reliability with Cronbach’s alpha of .74 and validity with good area under the ROC of 80% for the self-neglect short form. Box and whisker plots by substantiation decision revealed that standardized measures promote consistency since the majority of cases’ substantiation corresponded to a reasonable distribution of the ISO Matrix scores. Case reviews and caseworker interviews on outliers exposed how the standardized measures were used inconsistently, and this led to improvements in training. Cases considered to be anomalies occurred mostly because of deviation from instructions on completing the ISO Matrix provided during training at the start of the pilot demonstration, thus indicating additional training need. For example, the completion of ISO Matrix should be after investigations; a final determination of both ISO Matrix scores and final substantiation decisions should be based on the totality of information and caseworkers’ informed clinical judgment.
Although this study revealed mostly errors in the application of the intended procedures, it also provided caseworkers the opportunity to provide feedback on wording and procedures of the ISO Matrix that could be improved to adhere to caseworkers’ needs and preferences. Similar wording on questions for self-neglect and neglect confused caseworkers about which section to complete. Examples listed as part of the “other” item limited caseworkers’ tendency to endorse the item. On top of additional training to address these issues, changes were made to avoid further misunderstanding. For example, we added instructions on self-neglect items, such that they should be responded to when clients themselves are doing the harm, while neglect items should be responded to when an alleged abuser is present. Even though we considered dropping all the examples under “other,” some caseworkers commented that the examples helped their decision making to endorse the item. We decided to add frequently observed examples of self-neglect, like unpaid bills in financial self-neglect cases, under the “other” item to capture its occurrence. With the fewer items in the short form, identifying what indication of self-neglect was not captured in the short form and adding them as examples under the “other” item could accommodate this limitation of the ISO Matrix.
Moreover, the following circumstances suggested that training in APS practice and case management system was needed, instead of ISO Matrix training. The complex steps for adding additional abuse found during investigation seemed to cause errors in documenting substantiation decisions. Additionally, APS caseworkers are, at times, reluctant to make a confirmed substantiation without the client agreeing that the allegations are true and cooperating with the investigative process. However, based on the California APS program-wide training on consistency in determining findings of abuse, the caseworkers should have based confirmation on the evidence gathered even though the client refused to participate in the interview. New training was provided on the standards for consistency in determining findings of abuse to resolve this type of discrepancy. Caseworkers were instructed in the training that in order to meet the standards for confirmed substantiation, the evidence gathered during the investigation would need to show a greater than 51% likelihood that the abuse occurred. The unfounded substantiation is made if the evidence shows that there is a 49% or less likelihood that the abuse occurred. The inconclusive substantiation would be appropriate if neither of the above standards are met by the evidence that is gathered.
In this study, no valid ISO Matrix scores were found that contradicted the substantiation decision. Rather, the scores mostly revealed that caseworkers were not following the recommended procedures for the ISO Matrix for a variety of reasons. One positive observation was that most cases reviewed and interviewed in this study (outlier cases) were earlier cases when ISO Matrix initially launched. Since we hosted additional training sessions to clarify confusion, very few cases from later in the pilot demonstration period required review. One caseworker, when interviewed, attributed the new awareness of how the ISO matrix was to be completed to additional instruction from the supervisor. This indicated that as experience with the ISO Matrix procedure and findings determination accumulated, improvement was made.
Limitations
This practice study is limited in the small number of cases reviewed and caseworkers interviewed, and the fact that only self-neglect results were examined. Therefore, it is possible that our examples may be atypical, such that results may not be generalizable across APS cases more broadly nor across types of abuse. While it is a limitation that caseworkers create both the ISO Matrix scores and the substantiation decisions with no completely objective gold standard, this study illustrates how standardized measures may help the supervisors to examine, inform, and improve the consistency of this decision making.
Additionally, even though future harm can be modeled statistically using standardized data, it requires longitudinal data collection beyond the scope of this study. Since 45% of APS clients return to APS (Aurelien et al., Citation2019), future studies should examine why and who returns to APS to model future harm. Clients returning to APS sometimes indicate their willingness to seek help, so instead of always viewing reoccurrence as a negative outcome, the focus should be on reasons for returning to APS.
Conclusion
The self-neglect portion of the ISO Matrix used in this study illustrates the combined roles of caseworker judgment and standardized assessment in the APS case investigation. Caseworker’s clinical judgment, partly relying on observation, is essential to a complete assessment. Logically, ISO Matrix scores should correspond with substantiation decisions. For example, if high ISO Matrix scores are recorded with cases that were unfounded, this would be an opportunity to check the functioning of the system and the possibility of training needs. By charting ISO Matrix scores against substantiation decisions, supervisors can do these types of checks and reconciliations immediately and efficiently. Case management systems could even be designed to flag such occurrences automatically. Once the improved procedures are uniformly used properly, it may be possible for rough guidelines to be developed to help make even quicker and more accurate decisions. For instance, the median in the inconclusive category could become such a rough guideline for self-neglect. Those above it would be more like those confirmed and those below it would be more like the unfounded. Of course, tight guidelines or cutoff points are not reliable, but should merely serve as guides for consideration. A realistic scenario would be to refine zones of questionable scores to be considered that could help to reduce the number of inconclusive substantiations while supporting more accurate substantiation decisions that are ultimately made by the caseworker and supervisor. The standardized measures and scores enable a common vocabulary and metric for discussion and providing feedback to improve caseworkers’ performance as well as enabling caseworker input to improve the system. New caseworkers joining APS might benefit most from this common vocabulary. The standardized assessment tools are important for promoting consistency in determining findings, which has also become a statewide goal for APS in California.
This study is unusual insofar as it used standardized, quantitative measures to assess elder and vulnerable abuse and, subsequently, queried the caseworkers about their use of the measures in order to examine adherence to intended practice as well as opportunities for system improvement. The use of standardized practices promotes common definitions so that caseworkers, supervisors, clients, researchers, and other stakeholders are speaking the same language in terms of the types and severity of abuse. This systemization of language and scoring can facilitate communication throughout the identification, services, and outcome assessment process so that accountability is assured and that there is an empirical basis on which to base system improvement.
Declaration of Interest Statement
We have no conflict of interest to declare.
Acknowledgments
We would like to express our gratitude towards San Francisco and Napa Adult Protective Services, from caseworkers, supervisors, analysts, managers, directors, to deputy directors.
Data availability statement
The data that support the findings of this study are available from the corresponding author, P. L., upon reasonable request. Restrictions may apply to the availability of these data based on data usage agreement between Purdue University and San Francisco Adult Protective Services.
Additional information
Funding
References
- Aurelien, G., Beatrice, M., Cannizzo, J., Capehart, A., Gassoumis, Z., & Urban, K. (2019). Adult maltreatment data report 2018.The Administration for Community Living, U.S. Department of Health and Human Services. Retrieved from https://acl.gov/sites/default/files/programs/2019-12/2018%20Adult%20Maltreatment%20Report%20-%20Final%20v1.pdf
- Beach, S. R., Liu, P., DeLiema, M., Iris, M., Howe, M., & Conrad, K. J. (2017). Development of short form measures to assess four types of elder mistreatment: Findings from an evidence-based study of APS elder abuse substantiation decisions. Journal of Elder Abuse and Neglect, 29(4), 229–253. doi:10.1080/08946566.2017.1338171
- Connolly, M. T., Brandl, B., & Breckman, R. (2014). The elder justice roadmap: A stakeholder initiative to respond to an emerging health, justice, financial and social crisis. Retrieved from https://ncea.acl.gov/Library/Gov_Report/docs/EJRP_Roadmap.pdf
- Conrad, K. J., & Conrad, K. M. (2019). Abuser risk measure: Reports by alleged victims to adult protective services. The Gerontologist, 59(5), e403–e414. doi:10.1093/geront/gny091
- Conrad, K. J., Iris, M., & Liu, P. (2017). Elder abuse decision support system: Field test outcomes, abuse measure validation, and lessons learned. Journal of Elder Abuse and Neglect, 29(2–3), 134–156. doi:10.1080/08946566.2017.1314843
- Conrad, K. J., Iris, M., Ridings, J. W., Langley, K., & Anetzberger, G. (2010). Self-report measure of psychological abuse of older adults. The Gerontologist, 51(3), 354–366. doi:10.1093/geront/gnq103
- Conrad, K. J., Iris, M., Ridings, J. W., Langley, K., & Wilber, K. H. (2010). Self-report measure of financial exploitation of older adults. The Gerontologist, 50(6), 758–773. doi:10.1093/geront/gnq054
- Conrad, K. J., Liu, P., & Iris, M. (2019). Examining the role of substance abuse in elder mistreatment: Results from mistreatment investigations. Journal of Interpersonal Violence, 34(2), 366–391. doi:10.1177/0886260516640782
- Hosmer, D. W., & Lemeshow, S. (2000). Assessing the fit of the model. In Applied logistic regression (2nd ed., pp. 160–164). New York, NY: John Wiley and Sons.
- Iris, M., Conrad, K. J., & Ridings, J. W. (2014). Observational measure of elder self-neglect. Journal of Elder Abuse and Neglect, 26(4), 365–397. doi:10.1080/08946566.2013.801818
- Liu, P., Conrad, K. J., Conrad, K. M., & Hass, Z. (in progress). Telling the story of adult protective services: Piloting California adult protective services identification, services, and outcomes (ISO) Matrix. 09/30/17-09/29/19, U.S. Department of Health and Human Service, Administration for Community Living: 90EJIG0010-01–00
- Mosqueda, L., Wiglesworth, A., Moore, A., Nguyen, A., Gironda, M., & Gibbs, L. (2016). Variability in findings from adult protective services investigations of elder abuse in California. Journal of Evidence-Informed Social Work, 13(1), 34–44. doi:10.1080/15433714.2014.939383
- Park, K., Johnson, K., Flasch, S., & Bogie, A. (2010). Structuring decision in adult protective services. National Council on Crime and Delinquency. Retrieved from http://www.nccdglobal.org/sites/default/files/publication_pdf/focus-adultprotectiveservices.pdf
Appendix A.
Self-Neglect Items in the ISO Matrix
Self-neglect is an adult’s refusal or failure to perform essential self-care tasks. Self-neglect client has no designated person responsible for his/her care. Caseworker should observe whether self-neglect takes place.