Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Aug 1.
Published in final edited form as: Psychol Assess. 2015 Aug 31;28(5):523–538. doi: 10.1037/pas0000197

Development and Initial Psychometric Examination of the Home Safety and Beautification Assessment in Mothers Referred to Treatment by Child Welfare Agents

Brad Donohue 1, Michelle Pitts 1, Graig Chow 1, Stephen Benning 1, Arturo Soto-Nevarez 1, Christopher Plant 1, Daniel N Allen 1
PMCID: PMC4775469  NIHMSID: NIHMS710405  PMID: 26322799

Abstract

Unintentional injury is the leading cause of death among children, with approximately 45% of injuries occurring in and around the home. Rates of home injury are particularly high in the homes of caregivers who are referred for intervention services by child welfare agents. However, there are few validated methods of home safety assessment available. The Home Safety and Beautification Assessment (HSBA) was developed to assist intervention planning specific to home safety and appearance in a sample of 77 mothers who were referred to treatment by Child Welfare. Exploratory factor analysis of HSBA items indicated that safety and appearance factors emerged across rooms in the home, and internal consistencies were good. For each room, the sums of assessors' safety and appearance intervention priority item scores were correlated with the assessors’ global safety and appearance ratings of the entire home, respectively. The participants' overall room attractiveness scores were correlated with the assessors' overall room appearance intervention priority scores whereas the participants' ratings of overall room safety were not correlated with the assessors' overall room safety intervention priority scores. Participants' scores on the Abuse subscale of the Child Abuse Potential Inventory, personal income and education level were not associated with the assessors' home safety and appearance intervention priority ratings, suggesting the HSBA is assessing constructs that are distinct from child abuse potential and socioeconomic status. The results support the HSBA in a sample referred to treatment by Child Welfare agents.

Keywords: child abuse & neglect, child development, home safety and appearance, drug abuse, child welfare, child protective services, safety skills, treatment, intervention, assessment

Unintentional injury is the leading cause of death among children one to 19 years of age in the U.S., accounting for nearly 40 percent of deaths in this age group (CDC, 2012; Safe Kids Worldwide, 2008). As summarized by the CDC (2012) many of the unintentional injuries leading to death or causing serious and permanent disabilities are preventable. Yet each year almost nine million children and teenagers are treated in emergency departments for unintentional injuries, 225,000 are hospitalized, and 9,000 of these injuries prove fatal. The U.S. ranks among the worst of all high-income countries for child injury death rates. Unintentional fatal injuries have not declined at the same rate as other health conditions affecting children in the U.S., and the resulting cost of unintentional injuries to society, according to CDC (2012), is estimated at $87 billion each year.

The safety of a child’s home environment is critically important as the majority of unintentional injuries occur in the home (Danseco, Miller, & Spicer, 2000; Nagaraja et al., 2005; Phelan, Khoury, Kalkwarf, & Lanphear, 2005), with approximately 40 percent of deaths and 50 percent of nonfatal unintentional injuries among children under 14 years occurring in, or in close proximity to, the home (Safe Kids Worldwide, 2004). For instance, one out of every 180 toddlers is poisoned from improperly secured household medications (Schillie, Shehab, Thomas, & Budnitz, 2009). Child neglect is associated with an increased likelihood of childhood injuries due to home hazards (i.e., physical/environmental, emotional, medical, educational; Kaplan, Pelcovitz, & Labruna, 1999). In the United States more than 3 million intervention referrals are made for child abuse and neglect (CAN) annually, and these forms of maltreatment are often co-morbid (Department of Health and Human Services [DHHS], 2011). Child neglect is the most frequently indicated form of child maltreatment (HHS, 2012), and physical/environmental neglect, often specific to home environments, is the most prevalent form of child neglect, accounting for up to 57% of neglect cases (Sedlak & Broakhurst, 1996). Child neglect is associated with substance abuse, and both increase risk of home accidents (Murphy et al, 1991). Therefore, child neglect and drug abuse are prevalent health concerns that often result in home safety hazards.

The relevance of environmental factors in the home extends beyond safety, as other factors in the home may negatively impact the development of children. For example, organized homes that include books have been linked to positive outcomes in cognitive and language development of young children (Johnson et al., 2008), and household disarray and clutter have been indicated to be particularly deleterious to the functioning of socioeconomically distressed mothers (Deater–Deckard, Chen, Wang, & Bell, 2012). Other investigations have found the presence of objects (e.g., toys) influence neuronal plasticity and brain maturation during sensitive periods of development through stimulation (Lewis, 2004; Pollen et al., 2007; Gonzalez-Lima et al., 1994), and overcrowding (which may be influenced by excessive clutter) is associated with decreases in cognitive development in infants (Wachs & Chang, 1986). Hamid and Newport (1989) determined that color was found to influence mood in children, and Gifford (1988) found bright light (as compared with low light) and home-like décor (as compared with office-like décor) was associated with more intimate communication, perhaps due to more physiological arousal. In a retrospective study comparing hospital records of 46 patients who were assigned to view a brick wall or natural scene from their hospital room during their recovery from surgery (not randomly assigned although matched on relevant factors), Ulrich (1984) found those with the natural view had shorter postoperative hospital stays, received fewer negative evaluative comments in nurses' notes, and took fewer potent analgesics. Saxbe and Repetti (2010) found mothers who described their homes as messy, cluttered, sloppy, trashy/trashed, and unfinished evidenced greater levels of stress, depression, lower marital satisfaction, and flatter diurnal slopes of cortisol (which has been linked with adverse health outcomes), as compared with mothers who did not. Similar results were not found in fathers, suggesting mothers may be particularly vulnerable to negative effects due to home appearance, as compared with fathers. For more than a century professionals have pointed out the importance of teachers beautifying their classrooms through cleanliness, neatness, and decorating with borrowed portraits from the home environments of children to facilitate learning and a sense of “aesthetic culture” (Peaslee, 1880). Psychometrically validated measures of home appearance have been limited, although researchers have recently stressed the importance of assessing how physical home environmental features influence psychological states and development (Graham, Gosling, & Travis, 2015).

Standardized measures have been empirically developed to assess home cleanliness and safety. For instance, the Checklist for Living Environments to Assess Neglect (CLEAN; Watson-Perczel, Lutzker, Greene, & McGimpsey, 1988) may be used to examine objects not belonging in a certain area or room, including presence of matter that is decaying. The CLEAN is relatively limited in the number of cleanliness areas that can be assessed, but has good face validity, clinical utility and demonstrated inter-rater reliability. The CLEAN also permits assessment of home treatment foci that are more subtle than traditional home hazards, but nevertheless important to the development and safety of children (Lutzker, 1990).

Extant measures are available to assess home hazards. For instance, the Home Observation for Measurement of the Environment Inventory (HOME, Bradley & Caldwell, 1984; Caldwell & Bradley, 1984) assesses the child’s home learning environment. It involves both observation and semi-structured interview formats to examine safety and child-rearing activities in the home environment. Rating scales are developmentally sensitive, and although the original scale was limited to assessment of home environments for infants up to 6 years-old, it is now adapted for use in adolescents up to 14 years. Items were originally specific to interactions between caregivers and their children, discipline and emotional nurturing, intellectually stimulating activities, and the presence of affordable, age-appropriate toys and books. However, the HOME has expanded to a wider array of age-specific environmental domains (e.g., organization of the environment, learning tools, physical environment; Bradley, Caldwell, & Corwyn, 2003). HOME scores are significantly associated with socioeconomic status, suggesting the examination of home environment scales should consider diversity factors that may influence scoring, such as personal income.

The Home Accident Prevention Inventory Revised (HAPI-R; Lutzker, Bigelow, Doctor, & Kessler, 1998; Mandel, Bigelow, & Lutzker, 1998 ) may be utilized to assess hazards in homes as part of a home accident prevention program service, including type, quantity, and child accessibility of hazards (Barone, Green, & Lutzker, 1986). The HAPI-R is a checklist that includes 10 categories of common household hazards (e.g., suffocation, firearms, sharp objects, poisons, electrical, suffocation, falling, tripping, organic matter and allergens, activity restriction). Trained assessors use these categories to record the presence or absence of home hazards. The HAPI-R has demonstrated sufficient inter-rater reliability and content validity (Lutzker et al., 1998), and an earlier version of this scale demonstrated clinical utility in homes that are in need of safety hazard reduction (Barone et al., 1986; Tertinger, Greene, & Lutzker, 1984). One of the greatest assets of the HAPI-R is its direct applicability to treatment, as it is a vital component of successful treatment programs for children who are at-risk for child maltreatment, such as Project SafeCare (Gershater-Molko, Lutzker, & Sherman, 2003).

The Home Inventory of Dangers and Safety Precautions – 2 (HIDSP-2; Tymchuk, Lang, Dolyniuk, Berney-Ficklin, & Spitz, 1999) is a validated scale designed to assess frequency of dangers in homes, as well as associated safety precautions. This inventory assesses common factors in homes that may lead to unintentional childhood injuries. An advantage of this instrument is that it incorporates recommended precautions in the prevention of future danger. The scale includes 14 categories: fire, electrical, suffocation by ingested object, suffocation by mechanical object, fire arm/weapon, solid/liquid poisons, heavy object, sharp/pointed object, clutter, inappropriate edible, toy/animal, cooking, yard/outdoors, and general dangers. To design effective intervention or prevention programs, these investigators suggest that a functionally operationalized assessment is optimal (Mandel, Bigelow, & Lutzker, 1998; Tymchuk et al., 1999).

In existing home safety measures, assessors are trained to record the number of hazardous conditions in the home utilizing checklists that depict common safety hazards for children. Evaluators are encouraged to record notes that are relevant to each identified hazard. Although these measures have good face validity and clinical utility in prompting remediation strategies, psychometric support is warranted, particularly in regards to factorial, concurrent, and predictive validity. Moreover, items in the existing measures were not explicitly designed to assess intervention priority, and the relationship of endorsed items to child abuse potential has yet to be examined.

Current Study

This study was performed to develop a quantitative measure of home safety and appearance in a sample of mothers who were referred to intervention services by the local Department of Family Services (DFS) for child neglect and drug abuse. These mothers were predominately from low-income environments. We chose to examine home safety and appearance in this population as part of a larger treatment outcome study (Donohue et al., 2014) because low socio-economic status is the most reliable predictor of unintentional injury (Miller, Romano, & Spicer, 2000; Safe Kids Worldwide, 2004) and the children of these mothers are at-risk for unintentional injury and developmental delays due to maternal child neglect and substance abuse (Donohue, Romero & Hill, 2006). Therefore, the homes of these mothers were likely to evidence environmental hazards and lack positive developmental stimulation for their children.

The Home Safety and Beautification Assessment (HSBA) measure that was examined in this study is unique to other measures of home safety assessment because it incorporates three levels of analysis. Each item on the HSBA permits trained assessors to direct their attention to areas of the home that may be targeted for remediation. At the first level of analysis, trained assessors rate the extent to which home safety and appearance concerns are a priority for intervention in each room in the home. Responses to these HSBA items can be summed to create a total index of priority concerns for each room. At the second level of analysis, both trained assessors and parents complete overall room safety and overall appearance ratings to assess the characteristics of rooms holistically. At the third level of analysis, trained assessors complete global ratings of the home’s safety and appearance, taking into account all possible contributions to home safety and appearance of the home, even above and beyond those formally assessed using HSBA items.

It was hypothesized that the resulting measure would comprise two factors with adequate internal consistency. One factor was hypothesized to be associated with intervention priority specific to home hazards, and the other would be specific to intervention priority specific to home appearance. It was further hypothesized that room ratings for both safety hazard intervention priority and appearance intervention priority would predict global home safety and appearance intervention priority ratings, respectively. It was hypothesized that ratings by trained assessors and study participant would be positively correlated for both home safety and appearance intervention priority ratings. Lastly, it was hypothesized that there would be an association between the resulting HSBA factors and both child maltreatment potential and income because these factors are often present in parents who are referred to treatment by Child Protective Services (Donohue, Romero & Hill, 2006), and because household chaos has been influenced by socio-economic factors in mothers (Deater–Deckard, Chen, Wang, & Bell, 2012).

Method

Participants

One hundred and twenty-five mothers were referred to receive treatment of substance abuse and child neglect by the county’s Department of Family Services (DFS). Of these mothers, 94 were screened to initially meet study criteria (i.e., able to be contacted, interest in treatment offered, residing locally, reported drug use during past 4 months, referred for child neglect, not referred primarily for domestic violence, not receiving treatment, residing, or intention of residing, with the neglected child). Seventy-seven of these mothers provided study consent and were identified during baseline assessment to evidence a documented incident of child neglect and diagnosis of Substance Abuse or Dependence according to results obtained from the Structured Clinical Interview for DSM-IV (SCID-IV; Spitzer, Williams, Gibbon, & First, 1992) during baseline assessment.

Of the mothers who completed the baseline assessment and qualified for the study, the average age was 29.0 years (SD = 7.9 years, range = 18 – 49 years). The mean for the highest grade achieved was 11.3 (SD = 1.9, range = 5–16). Mothers reported that their personal monthly median income was $1,150 (range $0 to $12,500). On average, 1.6 minors resided in the home with the average age of the child being 3.8 years. Table 1 shows their ethnicity, marital status and neglect type.

Table 1.

Demographic Characteristics of Participants

Characteristic n %
Ethnicity
    Caucasian 39 50.6
    African American 16 20.8
    Hispanic 9 11.7
    American Indian 3 3.9
    Asian American 2 2.6
    Pacific Islander 2 2.6
    Other 6 7.8
Marital Status
    Single 35 45.5
    Cohabiting 26 33.8
    Married 15 19.5
    Unknown 1 1.3
Substantiated Neglect Type
    Drug use while pregnant 38 49.4
    Multiple types 11 14.3
    Lack of supervision 10 13.0
    Physical neglect 4 5.2
    Environmental neglect 3 3.9
    Emotional neglect 1 1.3
    Medical neglect 1 1.3
    Unclear 3 3.9
    Not provided 6 7.8
Open in a new tab

Measures

After study consent was obtained, a comprehensive battery of standardized assessment measures was administered in the homes of participating mothers. Measures utilized in the current study were selected from this battery (see below for description of relevant measures).

The Child Abuse Potential Inventory (CAPI; Milner, Gold, Ayoub, & Jacewitz, 1984) is a 160-item screening measure to assess the potential of parents to abuse their children. Although designed for physical abuse, the CAPI is valid for use in neglect populations because child abuse and neglect are frequently comorbid (Dong et al., 2004; McGee et al., 1995), and elevated CAPI Abuse scores have been shown to be a good predictor of future neglect (Ayoub & Milner, 1985). This study utilized the 77-item Abuse scale. Abuse scale scores range from 0 to 486, with higher scores indicating greater abuse potential. Scores at or above 215 are conservatively considered to reflect abuse potential. This inventory has demonstrated extensive psychometric support (Walker & Davies, 2010).

The Home Safety Beatification Assessment (HSBA) was inspired by the Home Accident Prevention Inventory – Revised (HAPI – R; Lutzker, Bigelow, Doctor, & Kessler, 1998). The intention was to develop a measure that could be used to assist concurrent enhancement of home safety and home beautification.

Assessors’ safety and appearance intervention priority items. In determining the individual HSBA items, a series of 6 focus groups was convened to initially generate items to assess home safety hazards and factors that negatively affect home appearance. Focus groups emphasized brainstorming analysis (Ritchie & Lewis, 2003). A moderator at the doctoral level directed discussion and identified key ideas (Krueger & Casey, 2000) while an assistant moderator at the doctoral level was responsible for recording comprehensive notes, utilizing a process facilitation approach with low content control and high process control (Millward, 1995). These moderator and assistant moderator were both experts in home-based assessment and treatment of child maltreatment and substance abuse. Other focus group members included five providers at the Bachelors level with experience in assessment and treatment involving parents in the child welfare system. Brainstorming was assisted by first reviewing the HAPI-R (Lutzker, Bigelow, Doctor, & Kessler, 1998) and HIDSP-2 (Tymchuk, Lang, Dolyniuk, Berney-Ficklin, & Spitz, 1999) to assist the origination of broad-based categories that were relevant to home safety. Categories were also determined from professional experience in the population under study. The resulting categories included toxins, electrical hazards, sharp objects, small objects, heavy/tipsy objects, food/nutrition, home access, air quality, cleanliness, and appearance. Comprehensive checklists of HSBA safety hazard items (e.g., frayed or exposed wires, heavy objects) were generated in brainstorming exercises for the kitchen, bedroom for which the child slept, bathroom used by the child, and family room. Items were generated for other rooms/areas in the home (i.e., dining room, garage, yard). However, these rooms were often missing in the homes of mothers who resided in low-income neighborhoods, and thus they were not a focus in the current study to permit greater applicability of the HSBA across participants.

A checklist of HSBA items was also similarly generated for items that were specific to appearance. We conceptualized appearance items as being relevant to cleanliness and aesthetic issues usually requiring minimal resources for remediation, and that albeit sometimes not a safety concern (torn/ripped furniture, absent décor) are often inexpensively managed to assist child development through visual stimulation and aesthetic culture. For instance, torn furniture may be sewn or patched, or children’s drawings can be put on empty walls and/or cover holes (see safety skills intervention used in Family Behavior Therapy, Donohue et al., 2015). Appearance items were originated to assist in the assessment of significant deficiencies that might result in illness, stress, negative mood, or inhibitions in child development (e.g., deteriorating rugs, dog feces on the kitchen floor, dishes in sink, food left out, clutter, appliances not working), and not to critique subjective choices in décor/fashion, suppress cultural expression, or evaluate the home to a standard that cannot be met due to financial restrictions.

Although most of the selected items appear across all rooms (e.g., exposed electrical wires), some items are unique to particular rooms. For example, the kitchen includes items that screen for the presence of sufficient nutritional foods while the bathroom includes an item specific to razors within reach of children. As can be seen in the Appendix, there are 40 safety items and 16 appearance items for the kitchen, 37 safety items and 16 appearance items for the bathroom, 40 safety items and 14 appearance items for the bedroom, and 40 safety items and 14 appearance items for the family room. Two caseworkers in a local child protective service agency substantiated that each of the generated safety items were relevant to safety, and each of the appearance items were relevant to cleanliness or aesthetics, respectively.

It was unanimously decided that the room checklists would be utilized to assist trained professionals in deriving overall room intervention priority rating scores for safety, and separately, appearance (i.e., cleanliness, aesthetics). Standardized protocols were developed to guide HSBA implementation. Trained raters tour each room and utilize the HSBA items to prompt examination of home hazards and appearance factors. Because the HSBA was developed to assist treatment providers in promoting safety and beautification of homes, all safety hazard and appearance items are rated on an intervention priority scale that is specific to remediation priority (0 = not present, 1 = present, no intervention priority, 2 = present, minimal intervention priority, 3 = present, moderate intervention priority, 4 = present, high intervention priority). Thus, a slightly worn and relatively clean carpet (indicated as an aesthetic item) may not be an intervention priority for a crawling infant (score = 1). However, a carpet that is worn to the extent that its rubber and glue are sticky to touch and have attracted pieces of food and dog feces, this carpet would be an intervention priority (perhaps a 4).

This scale of reference was chosen over severity-based scales (i.e., extremely hazardous) because we wanted to guide intervention specification and encourage motivation for growth through goal setting, while discouraging culpability through collaboration and positive perspective. There are guidelines for the assignment of each intervention priority rating that take into account the relative potential of harm posed by exposure to each item, as well as the developmental level of the child, to arrive at an appropriate rating. For instance, an uncovered electrical outlet has the potential for electrocution, but it poses a much greater risk to a toddler than to an adolescent. Considering the interaction between harm potential and developmental level, and because the HSBA was designed to assist immediate intervention when potential harm appeared impending, high priority items are selected for immediate intervention after HSBA implementation.

The assessors were initially trained to implement the HSBA during role-plays at the clinic. In doing so they were asked to independently rate various rooms (i.e., bathroom, waiting room, office) that had been set up to exemplify common safety hazards. When the assessors’ scores consistently matched their supervisor, they were permitted to administer the HSBA in practice cases that were not part of the current study while receiving supervision. When discrepancies between the supervisor and trainee were found, the trainee and supervisor discussed their scores and determined appropriate scores together. Although trainees and supervisors generally agreed on the majority of practice case items (prior to the current study), inter-rater reliability was not formally assessed.

Assessors’ overall room ratings for safety and appearance intervention priority. For each room in the home, assessors provide an overall safety intervention priority rating and an overall appearance intervention priority rating (0 = not present, 1 = present, no intervention priority, 2 = present, minimal intervention priority, 3 = present, moderate intervention priority, 4 = present, high intervention priority). For these overall intervention priority ratings, consideration was given to the number, and intervention priority, of HSBA safety and appearance items that were identified in each of the respective rooms. For instance, a room may receive a high intervention priority rating if only one hazard is identified, but this hazard poses an imminent threat of harm to the child (e.g., broken glass on the kitchen floor in the home of a toddler).

Assessors’ global home rating for safety and appearance intervention priority. For each home, assessors provide a global safety intervention priority rating and a global appearance intervention rating (0 = not present, 1 = present, no intervention priority, 2 = present, minimal intervention priority, 3 = present, moderate intervention priority, 4 = present, high intervention priority). For these global ratings, consideration was given to the number, and intervention priority, of HSBA safety and appearance items that were identified in all rooms. Similar to overall intervention priority scores that were derived for each of the rooms, the global intervention priority score for a home could reflect high intervention priority if one of the rooms included one hazard, but this hazard posed an imminent threat of harm to the child.

Participants’ overall room ratings for safety and attractiveness. For each room in the home, participants were instructed to provide ratings of overall safety and overall attractiveness without guidance or further instruction (1 = extremely unsafe to 6 = extremely safe; 1 = extremely unattractive to 6 = extremely attractive). Participants were not provided the HSBA to assist them with their overall scores, and they were not told the assessors’ ratings until the completion of this study. The rationale in doing so was to assist in determining how safe and attractive they believed their homes were without outside influences, such as comments by the trained raters’ or being prompted by the listed items. To ensure ratings between the trained assessors and participants were independently assessed in this study, the participants were instructed to place their completed ratings in an envelope without revealing scores to the trained assessor throughout the home tour.

Procedures

Department of Family Service (DFS) case workers were notified of the study, including study inclusion criteria through email and presentations at their facilities. Referrals were made by DFS caseworkers to participate in a treatment outcome study through telephone or fax. Upon receipt of the DFS referral, an intake specialist contacted the caseworker, and separately the participant, by telephone to determine if study inclusion criteria were met. Participants provided written consent to DFS to be contacted by the research team. Pre-qualifying participants were scheduled to complete informed consent and baseline assessment. All participants agreed to complete the HSBA, although trained assessors determined HSBA global home and overall room ratings without individual item scores for 5 participants due to time constraints, and in the administration of the HSBA one or two rooms were not assessed for 8 participants, usually because the particular type of room was not present (e.g., no living room) or less often because the person did not wish to assess the room. Participants were compensated for their time with a $50 gift card for use at local store for the pre-treatment assessment. The university’s Institutional Review Board approved all study procedures, a federal certificate of confidentiality was obtained to protect participants from releasing information due to court subpoenas, and no adverse events were determined in the study.

Results

Correlations between Assessors’ Summed Intervention Priority Scores for Individual Items in Each Room and Assessors’ Overall Room Intervention Priority Scores

For each room in the house we summed the assessors’ safety hazard intervention priority item scores into a single composite score. For each room we subsequently correlated the summed score with the assessor derived overall room safety intervention priority score. High positive correlations would support the validity of the individual safety item scores by suggesting the trained assessors were using HSBA items in each room to derive each of the overall room scores for safety intervention priority. Similar analyses were conducted for the summed assessors’ appearance intervention priority item scores and the assessors’ overall room appearance intervention priority scores for each of the respective rooms. The results of these correlations are presented in Table 2. All correlations were significant. The correlations between the summed individual item intervention priority scores for each room and overall room appearance intervention priority scores were uniformly larger than those between summed and overall safety intervention priority scores for rooms. Therefore, overall room ratings by the assessors appear to have been influenced by the appearance item scores more than safety item scores.

Table 2.

Correlations between the Assessors’ Safety and Appearance Intervention Priority Scores for Individual Items Summed for Each Room and the Assessors’ Overall Room Safety and Overall Room Appearance Intervention Priority Scores

Room Overall Room Safety & Sum of
Safety Scores for Items		 Overall Room Appearance & Sum of
Appearance Scores for Items
Kitchen .63*** .79***
Bathroom .61*** .77***
Bedroom .57*** .82***
Family Room .48*** .78***
Open in a new tab

Note.

***

p < .001. Each correlation coefficient shown is the correlation between the listed room’s overall safety or appearance intervention priority score and the sum of all HSBA safety or appearance intervention priority score for the corresponding room.

Factor Analysis of Assessors’ Overall Room Safety and Overall Room Appearance Intervention Priority Scores

To determine whether rooms or rating type (safety hazard vs. appearance) were more strongly associated, and to determine if overall room safety hazard and appearance intervention priority scores were distinct from one another, the overall room safety and appearance intervention priority ratings were subjected to factor analysis (see Table 3). Internal consistency reliabilities were computed for each resulting factor. Oblique (direct oblimin) rotation was utilized because safety hazards and appearance concerns were conceptualized as related constructs. A Kaiser-Meyer-Olkin measure of sampling adequacy (.86) and Bartlett’s test of sphericity (p < .001) met minimum standards indicative of the suitability for factor extraction.

Table 3.

Exploratory Factor Analysis Results for Overall Room Safety & Overall Room Appearance Scores

Factor
Item 1 2 h2 Mean SD
Bedroom Appearance .94 −.08 .80 2.03 1.19
Kitchen Appearance .81 .03 .69 1.71 1.13
Family Room Appearance .78 −.01 .60 1.46 1.15
Bathroom Appearance .71 .14 .64 1.72 1.11
Family Room Safety −.07 .76 .54 1.99 0.87
Bedroom Safety −.02 .69 .46 2.06 0.85
Bathroom Safety .09 .68 .54 2.00 0.94
Kitchen Safety .31 .55 .62 2.14 0.90
       Eigenvalues 4.48 1.18
       % Variance 56.0 14.8
Open in a new tab

Note. h2 = communality. Salient factor pattern matrix coefficients are in boldface. Factor 1 = Appearance subscale (presence and priority for remediation of appearance concerns). Factor 2 = Safety subscale (presence and priority for remediation of safety hazards).

Both the Kaiser criterion (eigenvalues greater than 1.0) and scree plot examination substantiated a two factor solution (eigenvalues 4.48, 1.18, 0.62, 0.51, 0.40, 0.36, 0.27, and 0.18). The first factor comprised the four appearance ratings of each room (Kitchen Appearance, Bathroom Appearance, Bedroom Appearance, and Family Room Appearance), and the second factor comprised the four safety ratings of each room (Kitchen Safety, Bathroom Safety, Bedroom Safety, and Family Room Safety). The two factors accounted for 70.8% of the variance in the model.

Communalities ranged from .46 to .80. All item loadings exceeded the minimum of .32 (Tabachnick & Fidell, 2001), with the lowest loading at .55 (Kitchen Safety) and no crossloadings. Cronbach’s alpha was .81 for the Appearance subscale and .90 for the Safety subscale, demonstrating a high level of internal consistency. The correlation between the Appearance and Safety subscales was .62 (p < .001). Results of the aforementioned analyses support that the assessors’ appearance ratings were distinct from the safety hazard ratings.

Relative Contribution of Assessors’ Overall Room Intervention Priority Scores to Assessors’ Global Home Intervention Priority Scores

To determine which overall room safety and appearance intervention priority scores might contribute to the global score reflecting the home’s safety and appearance, respectively, by the trained assessors, we conducted two multiple regression analyses. The first analysis involved regressing the global home safety intervention priority score on overall room safety intervention priority scores s for the Kitchen, Bathroom, Bedroom, and Family Room. The second analysis involved regressing the global home appearance intervention priority score on overall room appearance intervention priority scores for the Kitchen, Bathroom, Bedroom, and Family Room.

For the global home safety multiple regression analysis, the model was statistically significant, F(4, 63) = 23.5, p < .001. The assessors’ overall room intervention priority scores for the four rooms explained 59.8% (Adjusted R2 = .573) of the variance in their global home safety scores. As can be seen from Table 4, Kitchen Safety, Bedroom Safety, and Family Room Safety intervention priority scores were significant and positive predictors of global home safety intervention priority scores. Bathroom Safety scores were unrelated to Home Safety scores. The variance inflation factor (VIF) for each predictor was less than 2, which was substantially less than a VIF of 10 that indicates potentially problematic multicollinearity (Welsch, Belsley, & Kuh, 1980). Thus, global Home Safety scores were rated in a manner that was consistent with the overall room safety scores. However, it is notable that Bathroom Safety scores were unrelated to Home Safety scores, indicating that the bathroom did not uniquely contribute to the trained assessors’ ratings for Home Safety.

Table 4.

Regression Results for Assessors’ Overall Room Safety Intervention Priority Scores for Each Room in the Home Predicting Assessors’ Global Home Safety Intervention Priority Scores

Overall Rm. Safety Intervention
Priority Rating by Room:		 B (SE) β t
Kitchen Safety .26 (.09) .30 2.80**
Bathroom Safety .06 (.09) .07 0.67
Bedroom Safety .25 (.09) .27 2.68**
Family Room Safety .28 (.09) .31 3.02**
Open in a new tab

Note. R2 = .60 (p < .001).

**

p < .01.

For the Home Appearance multiple regression analysis, the model was statistically significant, F(4, 63) = 57.8, p < .001. The four overall room ratings explained 78.6% (Adjusted R2 = .772) of the variance in global home appearance scores. As can be seen in Table 5, Kitchen Appearance, Bathroom Appearance, Bedroom Appearance, and Family Room Appearance intervention priority scores were significant and positive predictors of global Home Appearance intervention priority scores. The largest VIF for a predictor in this analysis was 3.07, far below a level indicative of multicollinearity. The predictor variables explained a high percentage of the variance, indicating that the assessors’ global home appearance intervention priority scores were consistent with the overall room appearance intervention priority scores. Furthermore, all overall room ratings were associated with the global home appearance rating, indicating that each room contributed uniquely to the global home appearance rating. Overall room appearance ratings accounted for a higher proportion of the variance relative to the overall room safety ratings, indicating there is a relatively stronger association between overall room appearance intervention priority scores and the global home appearance intervention priority scores.

Table 5.

Regression Results for Assessors’ Overall Room Appearance Intervention Priority Room Rating Scores for Each Room in the Home Predicting Assessors’ Global Home Appearance Intervention Priority Rating Scores

Overall Rm. Safety Intervention
Priority Rating by Room:		 B (SE) β t
Kitchen Appearance .28 (.10) .27 2.97**
Bathroom Appearance .28 (.10) .27 2.98**
Bedroom Appearance .28 (.10) .28 2.74**
Family Room Appearance .21(.09) .20 2.28*
Open in a new tab

Notes. R2 = .79 (p < .001).

*

p < .05.

**

p < .01.

Correlations between Assessors’ and Participants’ Overall Room Scores

Assessor ratings of overall safety intervention priority scores for Kitchen, Bathroom, Bedroom, and Family Room were correlated with the corresponding participant ratings of safety. As originally coded, higher assessor ratings represented greater intervention priority whereas higher ratings of safety for participants indicate more unsafe conditions. Similarly, assessor ratings of overall Kitchen, Bathroom, Bedroom, and Family Room appearance priority for intervention were examined with participant attractiveness scores for the corresponding rooms. Higher ratings of appearance for trained assessors represented greater appearance intervention priority and for participants higher ratings indicated more unattractive conditions. To be consistent with the assessors’ overall room ratings, participant ratings were reverse coded prior to analysis so that high scores always indicated greater levels of concern. It was hypothesized that there would be positive correlations between the trained assessors’ safety and appearance intervention priority scores and the participants’ safety and attractiveness scores. The correlation coefficients are presented in Table 6. There were no significant correlations observed between the trained assessors’ priority for safety intervention and participants’ safety ratings for each room. Conversely, the trained assessor-rated overall room appearance intervention priority scores for the Kitchen, Bathroom, Bedroom, and Family Room were significantly correlated with the participant-rated appearance of the same rooms. In sum, though trained assessors and participants appeared to score the appearance of each room in the home similarly, their overall safety scores were not correlated for any rooms of the home.

Table 6.

Correlations between Assessors’ and Participants’ Overall Room Safety and Overall Appearance Scores for Each Room

Room Safety Appearance
Kitchen .15 .44***
Bathroom .16 .41***
Bedroom .11 .42***
Family Room .13 .31**
Open in a new tab

Note.

**

p < .01,

***

p < .001.

Correlations between the Mean of All Overall Room Safety and Appearance Intervention Priority Factor Scores Provided by the Assessors and the Participants’ Child Abuse Potential Scores

Correlation analyses were performed among Home Safety and Home Appearance factor scores (defined as the mean of all overall room safety intervention priority scores or overall room appearance intervention priority scores, respectively, as per trained assessors), and CAPI Abuse subscale scores ( M = 163, SD = 103, internal consistency = .92). These analyses revealed no significant relationships between the Home Safety factor and CAPI Abuse subscale scores, r(74) = .16, p = .158, or between the Home Appearance factor and CAPI Abuse subscale scores, r(74) = .21, p = .066. Likewise, independent samples t tests did not yield significant differences in the Home Safety and Home Appearance scores between participants with CAPI Abuse subscale scores in the range conservatively indicative of abuse (i.e., ≥215) and those who scored below the cutoff for child abuse, t(75)s < 1.4, ps > .17. Thus, child abuse potential was not correlated with HSBA factor scores, and participants who reported clinical levels of child abuse potential did not produce factor scores that were statistically different from participants who did not evidence significant levels of child abuse potential. These results suggest the HSBA is assessing factors that are distinct from child abuse potential.

Correlations between Assessors’ Home Safety and Home Appearance Intervention Priority Factor Scores and Education and Income

Correlations were conducted between Home Safety and Home Appearance factor scores and the participants’ highest year of education and log-transformed monthly income. Mother’s education was not correlated with Home Safety [r(74) = −.10, p = .407] or Home Appearance [r(74) = −.05, p = .665] factor scores. Similarly, income was not correlated with Home Safety [r(75) = −.05, p = .645] or Home Appearance [r(75) = .05, p = .654] factor scores. This pattern indicates that socioeconomic status was not associated with home safety or appearance, as assessed by trained assessors utilizing the HSBA in this sample.

Discussion

This study was conducted to develop, and examine an instrument that was constructed to assess home safety and home appearance intervention priority. The HSBA was formatted to directly apply to intervention, emphasizing observationally specified item content and intervention priority ratings to facilitate positive feedback during treatment planning. The HSBA builds upon well-established measures (e.g., HAPI-R and HIDSP-2) by systematically prompting specific potential safety hazards and appearance concerns that are particular to common rooms in the home. Responses to this scale may facilitate identification of a wide range of potential safety hazards and appearance factors warranting intervention.

Clinical Applications of the HSBA

The real-world context in which HSBA scores were evaluated in this study is important because it demonstrates the external validity of this measure’s scores. Indeed, participants in this research were referred by DFS caseworkers due to illicit behavior (i.e., drug use, child neglect), potentially leading them to be relatively guarded as neglect charges are often brought against parents due to safety hazards (Ewigman, Kivlahan, & Land, 1994; Peterson & Brown, 1994). Moreover, many of these parents demonstrated a wide array of safety and appearance concerns warranting intervention, as consistent with the nature of their referral to treatment (Donohue, Romero, & Hill, 1996); suggesting the study population was relevant to HSBA development. Within these circumstances, this study appears to have psychometrically supported aspects of the HSBA.

For each room in the home, the sum of trained assessors’ HSBA safety intervention priority item scores was associated with the trained assessors’ overall room safety intervention priority score. Similar results were evidenced for appearance. These results are consistent with the contention that trained assessors examined the individual items to derive overall scores. In this way, the overall room ratings may be used to accurately and economically summarize intervention priorities, while the HSBA intervention priority item scores can be quickly examined to pinpoint intervention needs. For instance, treatment providers may be trained to utilize HSBA scores to first determine specific areas in the home that are relatively safe, clean and beautiful so they can descriptively praise parents for these accomplishments. The HSBA scores may subsequently be used to prioritize specific areas in the home for remediation, cleanliness, and beautification in collaboration with parents. In accomplishing these enhancements, family members foster a sense of pride that is reinforced by treatment providers. The reliable implementation of safety skills training, utilizing HSBA scores, is delineated in Donohue et al. (2015), and its application to intervention implementation is a relative strength of this measure. Exploratory factor analysis indicated that the trained assessors discriminated between safety and appearance items, thus appreciating subtle distinctions in the assessment format guidelines. The exploratory factor analysis also demonstrated this instrument’s factorial validity.

Relationships among Home Safety, Home Appearance and Other Factors

Except in the child’s bathroom, overall room safety and appearance scores predicted the global home safety and global appearance scores, respectively. These regressions suggested all but one of the overall room ratings were influencing global home ratings, further supporting the HSBA’s validity. Although it is difficult to determine why the assessors’ scores for intervention priority for the bathroom were not associated with their global home safety scores, it does appear that potential safety and appearance concerns for bathrooms may be functionally distinct from other rooms in the house.

Interestingly, appearance and attractiveness scores between trained assessors and participants, respectively, were significantly associated with one another for each room in the home. This result suggests the HSBA is reliable when trained assessors and untrained independent raters assess its appearance items, which may facilitate acceptance of interventions that are designed to enhance home appearance. However, for all rooms in the house, overall safety concerns were rated differently between trained assessors and the participants in this study. In understanding these results it is important to consider that participants may overlook safety hazards in their homes when compared with professionals who are prompted to prioritize potential safety hazards utilizing the HSBA. This explanation would suggest the HSBA may have practical utility in the identification and subsequent remediation of home safety hazards that may be overlooked by caregivers who may benefit from home safety interventions. It may also be that the HSBA influences trained assessors to be overly sensitive to prioritizing situations as requiring safety remediation when these situations are not unsafe at all, or unlikely to be hazardous. Indeed, a strong draft under a door might be identified as a priority for safety intervention if the assessment occurs when the weather is 20 degrees below freezing (i.e., may influence a child to become sick), while the same draft may be overlooked in 70 degree weather. In these cases the assessors must be trained to consider all factors that may influence safety in the future, and always solicit the participating family members in the process of intervention planning. In this way, false positives (identifying an item as hazardous when in fact it is not) are benignly cautionary. Therefore, the results indicate the safety intervention priority scores of trained assessors may yield false positives, which require trained assessors to be sensitive to various environmental circumstances, and demonstrate flexibility in the management of results. Moreover, it is important to emphasize that the results of this study do not support the use of HSBA scores to substantiate child abuse. Alternatively, these results might indicate at least some of the participants are more unaware of safety hazards than trained assessors, suggesting these participants may benefit from safety skill interventions.

There is some support to suggest HSBA scores in this sample were not influenced by the participants’ income, educational level, or child abuse potential. These results also suggest the item pool may be appropriately assessing a broad array of items that may be appropriate to families in relatively low income levels of income and educational level. That said, it is important to interpret income results with caution due to restricted range (i.e., most of the participants were from low-income backgrounds). Nevertheless, assessors were trained to appreciate the concerns of mothers who were disproportionately represented from low-income backgrounds and sometimes mandated to receive assessment and treatment. In this way assessors were trained to be respectful and considerate at all times. For instance, assessors queried participants for permission to open kitchen cabinets, avoided patronization, rated items without judgment, actively listened, empathized and solicited solutions to concerns. Nevertheless, it is very possible some of the HSBA items may reflect economic disadvantage.

Limitations and Future Directions

Overall, the present study aligns well with the current direction of child maltreatment research and intervention that proposes to develop comprehensive assessment measures of home safety hazards (Gershater-Molko, Lutzker, & Sherman, 2003). In this regard, the HSBA has broad-based applications. For instance, in its most basic form, the HSBA can be used to assess home safety hazards and appearance issues in at-risk groups, such as child neglect and substance abuse. However, it should be emphasized that not having a normative control group comparison in the current study was a limitation. Indeed, the HSBA should be examined in other populations that are at-risk for unintentional injury (e.g., parents with intellectual disabilities), and compare these at-risk populations with parents who are not at-risk for home safety hazards and home appearance concerns. It would be expected that at-risk populations would demonstrate higher intervention priority scores on the HSBA than non-risk populations. However, if the scores were similar, this might suggest the HSBA is relatively robust. Investigators should also examine inter-rater reliability of the HSBA by comparing HSBA scores of independent trained raters. Indeed, this was a significant limitation in the present study.

The relatively low number of participants in the current study is also a limitation, and the sample was limited to mothers. These limitations may affect the generalizability of study findings and warrant further examination of the HSBA in fathers and other caregivers. Lastly, it is important that child welfare treatment programs attempt to utilize evidence-supported safety skill assessments and treatment components in their treatment planning. The HSBA, and other home safety assessments (e.g., HOME, HAPI-R, CLEAN, HID SP-2), may prove useful in helping to identify caregivers who could profit from safety skill intervention services that have been indicated to prevent unintentional injuries, such as Project 12 Ways and SafeCare (e.g., Metchikian et al., 1999), which are model intervention approaches in this regard. The HAPI-R that is used in Project SafeCare and the HSBA utilized in Family Behavior Therapy (Donohue et al., 2015) are similar in that family members can be assisted in generating solutions to identified issues that are assessed to require intervention (Metchikian et al., 1999). The HSBA may additionally be used to inform treatment planning through intervention prioritization scores (as mentioned above).

As a concluding remark, it is important to indicate that information gathered through the HSBA may prove useful to social workers, counselors, psychologists, and court liaisons by objectively demonstrating that although there is no evidence that home safety hazards and home appearance are associated with the severity of child abuse potential, home safety and appearance factors are certainly relevant in caregivers who are indicated to neglect their children and abuse substances.

Acknowledgements

This study was supported with a grant awarded to Brad Donohue from the National Institute on Drug Abuse (1R01DA020548-01A1).

Appendix

Home Safety and Beautification Assessment

graphic file with name nihms710405f1.jpg

graphic file with name nihms710405f2.jpg

graphic file with name nihms710405f3.jpg

graphic file with name nihms710405f4.jpg

graphic file with name nihms710405f5.jpg

graphic file with name nihms710405f6.jpg

graphic file with name nihms710405f7.jpg

References

  1. Ayoub CC, Milner JS. Failure to thrive: Parental indicators, types, and outcomes. Child Abuse and Neglect. 1985;9:491–499. doi: 10.1016/0145-2134(85)90058-4. [DOI] [PubMed] [Google Scholar]
  2. Barone VJ, Greene BF, Lutzker JR. Home safety with families being treated for child abuse and neglect. Behavior Modification. 1986;10:93–114. doi: 10.1177/01454455860101006. [DOI] [PubMed] [Google Scholar]
  3. Bradley RH, Caldwell BM. The HOME inventory and family demographics. Developmental Psychology. 1984;20:315–320. [Google Scholar]
  4. Bradley RH, Caldwell BM, Corwyn RF. The child care HOME Inventories: Assessing the quality of family child care homes. Early Childhood Research Quarterly. 2003;18:294–309. [Google Scholar]
  5. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. National action plan for child injury prevention. 2012 Retrieved from http://www.cdc.gov/safechild/pdf/National_Action_Plan_for_Child_Injury_Prevention.pdf.
  6. Caldwell BM, Bradley RH. Administration manual, revised edition: Home observation for measurement of the environment. Little Rock, AR: University of Arkansas; 1984. [Google Scholar]; Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. National action plan for child injury prevention. 2012 Retrieved from http://www.cdc.gov/safechild/pdf/National_Action_Plan_for_Child_Injury_Prevention.pdf.
  7. Danseco ER, Miller TR, Spicer RS. Incidence and costs of 1987–1994 childhood injuries: Demographic breakdowns. Pediatrics. 2000;105:e27–e27. doi: 10.1542/peds.105.2.e27. [DOI] [PubMed] [Google Scholar]
  8. Deater-Deckard K, Chen N, Wang Z, Bell MA. Socioeconomic risk moderates the link between household chaos and maternal executive function. Journal of Family Psychology. 2012;26:391–399. doi: 10.1037/a0028331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dong M, Anda RF, Felitti VJ, Dube SR, Williamson DF, Thompson TJ, Giles WH. The interrelatedness of multiple forms of childhood abuse, neglect, and household dysfunction. Child Abuse and Neglect. 2004;28:771–784. doi: 10.1016/j.chiabu.2004.01.008. [DOI] [PubMed] [Google Scholar]
  10. Donohue B, Azrin N, Bradshaw K, Van Hasselt V, Cross C, Urgelles J, Allen DN. A controlled evaluation of family behavior therapy in concurrent child neglect and drug abuse. Journal of Consulting and Clinical Psychology. 2014;82:706–720. doi: 10.1037/a0036920. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Donohue B, Romero V, Hill H. Treatment of co-occurring child maltreatment and substance -abuse. Aggression and Violent Behavior: A Review Journal. 2006;11:626–640. [Google Scholar]
  12. Ewigman B, Kivlahan C, Land G. The Missouri child fatality study: Underreporting of maltreatment fatalities among children under five years of age. Pediatrics. 1993;91:330–337. [PubMed] [Google Scholar]
  13. Gershater-Molko RM, Lutzker JR, Sherman JA. Assessing child neglect. Aggression and Violent Behavior. 2003;8:563–585. [Google Scholar]
  14. Gifford R. Light, décor, arousal, comfort, & communication. Journal of Environmental Psychology. 1988;8:177–189. [Google Scholar]
  15. Gonzalez-Lima F, Ferchmin PA, Eterovic VA, Gonzalez-Lima EM. Metabolic activation of the brain of young rats after exposure to environmental complexity. Developmental Psychobiology. 1994;27:343–351. doi: 10.1002/dev.420270603. [DOI] [PubMed] [Google Scholar]
  16. Graham LT, Gosling SD, Travis CK. The Psychology of Home Environments: A Call for Research on Residential Space Perspectives on Psychological Science. 10:46–356. doi: 10.1177/1745691615576761. [DOI] [PubMed] [Google Scholar]
  17. Hamid PN, Newport AG. Effect of colour on physical strength and mood in children. Perceptual and Motor Skills. 1989;69:179–185. doi: 10.2466/pms.1989.69.1.179. [DOI] [PubMed] [Google Scholar]
  18. Johnson AD, Martin A, Brooks-Gunn J, Petrill SA. Order in the house! Associations among household chaos, the home literacy environment, maternal reading ability, and children's early reading. Merrill-Palmer Quarterly. 2008;54:445–472. doi: 10.1353/mpq.0.0009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kaplan SJ, Pelcovitz D, Labruna V. Child and adolescent abuse and neglect research: A review of the past 10 years. part 1: Physical and emotional abuse and neglect. Journal of American Academic Child and Adolescent Psychiatry. 1999;38:1214–1222. doi: 10.1097/00004583-199910000-00009. [DOI] [PubMed] [Google Scholar]
  20. Krueger RA, Casey M. Focus groups: A practical guide for applied research. 3rd ed. Thousand Oaks, CA: Sage; 2000. [Google Scholar]
  21. Lewis MH. Environmental complexity and central nervous system development and function. Mental Retardation and Developmental Disabilities Research Reviews. 2004;10:91–95. doi: 10.1002/mrdd.20017. [DOI] [PubMed] [Google Scholar]
  22. Lutzker JR. Behavioral treatment of child neglect. Behavior Modification. 1990;14:301–315. doi: 10.1177/01454455900143005. [DOI] [PubMed] [Google Scholar]
  23. Lutzker JR, Bigelow KM, Doctor RM, Kessler ML. Safety, health care, and bonding within an ecobehavioral approach to treating and preventing child abuse and neglect. Journal of Family Violence. 1998;13:163–185. [Google Scholar]
  24. Mandel U, Bigelow KM, Lutzker JR. Using video to reduce home safety hazards with parents reported for child abuse and neglect. Journal of Family Violence. 1998;13:147–162. [Google Scholar]
  25. McGee RA, Wolfe DA, Yuen SA, Wilson SK, Carnochan J. The measurement of maltreatment: A comparison of approaches. Child Abuse and Neglect. 1995;19:233–249. doi: 10.1016/0145-2134(94)00119-f. [DOI] [PubMed] [Google Scholar]
  26. Metchikian KL, Mink JM, Bigelow KM, Lutzker JR, Doctor RM. reducing home safety hazards in the homes of parents reported for neglect. Child & Family Behavior Therapy. 1999;21:23–34. [Google Scholar]
  27. Miller TR, Romano EO, Spicer RS. The cost of childhood unintentional injuries and the value of prevention. Future of Children. 2000;10:137–163. [PubMed] [Google Scholar]
  28. Millward LJ. Focus groups. In: Breakwell GM, Hammond S, Fife-Schaw C, editors. Research methods in psychology. London: Sage Ltd; 1995. pp. 274–292. [Google Scholar]
  29. Milner JS, Gold RG, Ayoub C, Jacewitz MM. Predictive validity of the child abuse potential inventory. Journal of Consulting and Clinical Psychology. 1984;52(5):879. doi: 10.1037//0022-006x.52.5.879. [DOI] [PubMed] [Google Scholar]
  30. Murphy JM, Jellinek M, Quinn D, Smith G, Poitrast FG, Goshkom M. Substance abuse and serious child mistreatment: Prevalence, risk, and outcome in a court sample. Child Abuse & Neglect. 1991;15:197–211. doi: 10.1016/0145-2134(91)90065-l. [DOI] [PubMed] [Google Scholar]
  31. Nagaraja J, Menkedick J, Phelan KJ, Ashley P, Zhang X, Lanphear BP. Deaths from residential injuries in US children and adolescents, 1985–1997. Pediatrics. 2005;116:454–461. doi: 10.1542/peds.2004-1415. [DOI] [PubMed] [Google Scholar]
  32. Peaslee JB. Aesthetic culture. Journal of Education. 1884;19:23. [Google Scholar]
  33. Peterson L, Brown D. Integrating child injury and abuse-neglect research: common histories, etiologies, and solutions. Psychological Bulletin. 1994;116(2):293. doi: 10.1037/0033-2909.116.2.293. [DOI] [PubMed] [Google Scholar]
  34. Phelan KJ, Khoury J, Kalkwarf H, Lanphear B. Residential injuries in US children and adolescents. Public Health Reports. 2005;120:63. doi: 10.1177/003335490512000111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Pollen AA, Dobberfuhl AP, Scace J, Igulu MM, Renn SP, Shumway CA, Hofmann HA. Environmental complexity and social organization sculpt the brain in the Lake Tanganyikan cichlid fish. Brain, Behavior and Evolution. 2007;70:21–39. doi: 10.1159/000101067. [DOI] [PubMed] [Google Scholar]
  36. Ritchie J, Lewis J, editors. Qualitative research practice: A guide for social science students and researchers. London: Sage Ltd; 2003. [Google Scholar]
  37. Safe Kids Worldwide. Childhood Injury Fact Sheet. Washington, DC: National Safe Kids Campaign; 2004. [Google Scholar]
  38. Safe Kids Worldwide. Report to the nation: Trends in unintentional childhood injury mortality and parental views on child safety. 2008 Retrieved from http://www.safekids.org/assets/docs/ourwork/research/research-report-safe-kids-week-2008.pdf.
  39. Saxbe DE, Repetti R. No place like home: Home tours correlate with daily patterns of mood and cortisol. Personality and Social Psychology Bulletin. 2010;36:71–81. doi: 10.1177/0146167209352864. [DOI] [PubMed] [Google Scholar]
  40. Schillie SF, Shehab N, Thomas KE, Budnitz DS. Medication overdoses leading to emergency department visits among children. American Journal of Preventive Medicine. 2009;37:181–187. doi: 10.1016/j.amepre.2009.05.018. [DOI] [PubMed] [Google Scholar]
  41. Sedlak AJ, Broadhurst DD. Third national incidence study of child abuse and neglect. Washington, DC: U. S. Government Printing Office; 1996. [Google Scholar]
  42. Spitzer RL, Williams JB, Gibbon M, First MB. The Structured Clinical Interview for DSM-III-R (SCID) I: history, rationale, and description. Archives of General Psychiatry. 1992;49:624. doi: 10.1001/archpsyc.1992.01820080032005. [DOI] [PubMed] [Google Scholar]
  43. Springer KW, Sheridan J, Kuo D, Carnes M. Long-term physical and mental health consequences of childhood physical abuse: Results from a large population-based sample of men and women. Child Abuse & Neglect. 2007;31:517–530. doi: 10.1016/j.chiabu.2007.01.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Tabachnick BG, Fidell LS. Using Multivariate Statistics. Boston: Allyn and Bacon; 2001. [Google Scholar]
  45. Tertinger DA, Greene BF, Lutzker JR. Home safety: Development and validation of one component of an ecobehavioral treatment program for abused and neglected children. Journal of Applied Behavior Analysis. 1984;17:159–174. doi: 10.1901/jaba.1984.17-159. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Tymchuk AJ, Lang CM, Dolyniuk CA, Berney-Ficklin K, Spitz R. The home inventory of dangers and safety precautions-2: Addressing critical needs for prescriptive assessment devices in child maltreatment and in healthcare. Child Abuse & Neglect. 1999;23(1):1–14. doi: 10.1016/s0145-2134(98)00107-0. [DOI] [PubMed] [Google Scholar]
  47. Ulrich RS. View through a window may influence recovery from surgery. Science. 1984;224:420–421. doi: 10.1126/science.6143402. [DOI] [PubMed] [Google Scholar]
  48. U.S. Department of Health and Human Services, Administration for Children and Families, Children’s Bureau, Child Welfare Information Gateway. Child Maltreatment 2011. 2012 Retrieved from http://www.acf.hhs.gov/programs/cb/research-data-technology/statistics-research/child-maltreatment.
  49. U.S. Department of Health and Human Services, Administration for Children and Families, Children’s Bureau, Child Welfare Information Gateway. Definitions of child abuse and neglect. 2011 Retrieved from https://www.childwelfare.gov/systemwide/laws_policies/statutes/define.pdf.
  50. Wachs TD, Chan A. Specificity of environmental action, as seen in environmental correlates of infants’ communication performance. Child Development. 1986;57:1464–1474. [Google Scholar]
  51. Walker CA, Davies J. A critical review of the psychometric evidence base of the Child Abuse Potential Inventory. Journal of Family Violence. 2010;25:215–227. [Google Scholar]
  52. Watson-Perczel M, Lutzker JR, Greene BF, McGimpsey BJ. Assessment and modification of home cleanliness among families adjudicated for child neglect. Behavior Modification. 1988;12:57–81. doi: 10.1177/01454455880121003. [DOI] [PubMed] [Google Scholar]
  53. Welsch Ra, Belsley DA, Kuh EA. Regression diagnostics: identifying influential data and sources of collinearity. Ipswich, MA: Wiley; 1980. [Google Scholar]

RESOURCES