Skip Navigation

Implementing SafeCare®

Yes (SafeCare Augmented only)

Implementation Experiences

Last Updated

November 2013


Summary of Sources

Information in this section is based on studies included in the HomVEE review. For Project 12-Ways/SafeCare, we reviewed nine studies, including three randomized controlled trials (RCTs) and one single subject study and five standalone implementation studies. (Please see the study database for a list of the studies.)


Characteristics of Model Participants

Across two of the studies reviewed, the average maternal age ranged from 26 to 28 years. Across three studies, the average child’s age ranged from 3 to 5 years.

In three studies, only mothers participated; another study included fathers. Across 4 of the studies, from 19 to 255 families participated in SafeCare, and another study reported that more than 2,000 families had enrolled in one statewide trial.

Families enrolled in the program in four of the studies were racially and ethnically diverse. The percentage of program participants that were African American ranged from 7 to 83 percent. The percentage of program participants that were white ranged from 8 to 38 percent. The percentage of participants that were Hispanic ranged from 6 to 64 percent.

One of the studies reported that 50 percent of the program participants were unemployed; 63 percent had a high-school diploma or general education diploma (GED); and 38 percent received Temporary Assistance for Needy Families (TANF). Another study reported that families’ average estimated annual incomes were approximately $18,000 (year of annual income was not documented).

Participation was voluntary in four studies.


Location and Setting

One of the studies took place in California; two studies took place in Georgia; and two studies took place in South Bend, Indiana, and Kansas City (state not specified). In another study, sites were located in six states: California, Colorado, Indiana, Kansas, Oklahoma, and Washington.

Two studies described the agency that implemented SafeCare. In one study, the program was implemented by a local, nonprofit social service organization. In another study, state agencies with active contracts to provide family preservation services in one or more counties were trained to deliver SafeCare.


Staffing and Supervision

Five of the studies reported that SafeCare employed service providers from diverse backgrounds, such as graduate students, undergraduate research assistants, nurses, and caseworkers. In one study, the program manager was a graduate student who hired and supervised behavioral parenting specialists. Most of these specialists had recently received their bachelor’s degrees; others were graduate students in psychology, behavior analysis, counseling, education, social work, and other related fields.

In one study, behavioral parenting specialists had to have at least a bachelor’s degree and one year of human service experience. They were supervised by a licensed psychologist.


Model Components

Six studies reported that the program included only home visiting. According to four of the studies, home visits encompassed baseline and follow-up assessments, observations, and trainings with parents. Three studies described the training component. Trainings covered three modules: health, home safety, and parent-infant or parent-child interactions. During the trainings, home visitors explained the rationale for each concept, modeled the concepts for the parents, encouraged the parents to practice the steps, and then provided feedback. One study reported that each module was intended to be covered over five weeks. Two studies that focused on SafeCare’s parent-child interaction module, Planned Activities Training (PAT), reported that PAT consisted of five to seven sessions conducted approximately weekly.

One study stated that the full service period was intended to last 18 to 20 weeks, depending on parents’ progress in meeting skill-based criteria for each module. The study also stated that home visits were intended to occur weekly and last one to two hours.

Two studies described the PAT module. PAT seeks to teach parenting strategies that will improve parent-child interactions and prevent challenging child behavior. Parenting strategies include establishing limits and rules, planning and explaining activities in advance, and employing positive interaction skills.

Four studies reviewed reported that families enrolled in SafeCare were assessed in various ways, such as via questionnaires, direct observations of parent-child interactions, and direct behavioral assessment of parents’ health care skills as well as home safety. The assessments were conducted at the start and end of each of the three program modules (home safety, infant and child health care, and bonding and stimulation).

One study mentioned that materials were available in Spanish. Another study reported that assessments were administered in English or Spanish according to parents’ preferences.


Model Adaptations or Enhancements

Five of the studies reviewed were based on adaptations or enhancements to SafeCare or a variant of the SafeCare model.

Two studies examined a cellular phone enhancement to PAT (CPAT). In addition to receiving PAT training sessions via home visits, CPAT participants received a cell phone at their initial visit to facilitate contact with their PAT home visitor. CPAT guidelines recommended that home visitors call participants between home visits (approximately once per week); send twice-daily text messages (approximately 10 messages per week); and encourage parents to call for advice, information, or support. The timing, duration, and frequency of calls was guided by parent preferences, although the home visitor attempted to ensure that no more than five days passed without a home visit or telephone contact. During telephone calls, home visitors asked open-ended questions about how the parents were doing generally and with their practice of PAT strategies, as well as about their interactions with their children and their children’s behavior. Text messages reinforced the parents’ use of PAT strategies. Home visitors individualized text messages based on the focus of their recent home visits.

In the third study, SafeCare was augmented with Motivational Interviewing plus additional training for home visitors to identify and respond to imminent child maltreatment and risk factors of substance abuse, depression, and intimate partner violence. The adaptation, called SafeCare Augmented, was further modified for high-risk, rural communities. Monitors certified by the national developers trained and observed home visitors for model fidelity, and a member of the Motivational Interviewing Network of Trainers provided pre- and in-service motivational interviewing training. Local experts in intimate partner violence, substance abuse, and mental health provided ongoing case consultation.

In the fourth study, families enrolled in Project 12-Ways, a precursor to SafeCare, received a supplemental home safety education package designed to reduce home safety hazards to children under 5 years of age. The program covered up to five topics from the Home Accident Prevention Inventory (HAPI): poisoning by solids and liquids; suffocation by mechanical objects; fire and electrical hazards; suffocation by ingested objects; and firearms. The enhancement encompassed: (1) an audio slideshow to instruct parents on relevant home safety hazards and model ways of making the hazards inaccessible, such as installing safety latches; (2) self-feedback stickers, designed as a white circle with a red circumference and diameter representing “No,” to place on storage areas accessible to children; (3) home safety accessories including safety latches, electrical and switch plates, and electrical tape; and (4) a home safety review manual consisting of images from the slideshow and guidelines for using the “No” stickers and safety accessories. Counselors from Project 12-Ways conducted two home visits per topic and unannounced follow-up home safety checks approximately every 2 to 3 weeks until families completed Project 12-Ways. Counselors assessed home safety using the HAPI during each visit.

The fifth study adapted the UCLA Parent-Child Health and Wellness Project, which is based on SafeCare, for an Australian context. Several adaptations were necessary. Language was changed to reflect Australian usage, such as changing the word “crib” to “cot.” The content was modified to reflect cultural differences, including making a section on firearms optional. In addition, all project materials were checked with Australian agencies and altered to conform to Australian standards. For example, the original curriculum teaches the use of sunscreen to prevent sunburn whereas Australia recommends covering up and keeping out of the sun between 11 a.m. and 3 p.m. Finally, the project materials were checked by a small advisory group of parents with intellectual disabilities for readability and presentation. On their advice, color and graphics were added and language simplified to increase accessibility and enhance interest for parents with low literacy skills.



None of the studies reviewed included information about (1) how many home visits program participants actually received, (2) the length of the visits, or (3) the average duration of participants’ enrollment in SafeCare.


Fidelity Measurement

In one study, fidelity was monitored by the program manager consistently attending family sessions to conduct observations or by reviewing videotapes of home visit sessions. Individual feedback was given to the home visitor and any deviation from the model was discussed. The program manager would then monitor the situation as needed. In another study, a SafeCare trainer or certified on-site coach monitored home visitors who had recently completed training as they implemented SafeCare with their first family. All sessions were audio-recorded and at least two of the initial four sessions were observed live. SafeCare trainers and on-site coaches monitored fidelity using standardized checklists with approximately 30 items pertaining to expected home visiting behaviors.



None of the studies reviewed included information about the costs of implementing SafeCare.


Lessons Learned

Several lessons were learned about implementation as described in two of the studies reviewed: (1) fidelity monitoring is crucial to the achievement of successful implementation; (2) clients prefer in-person home visits, rather than a mix of video and in-person trainings; and (3) at-risk clients are more responsive to direct-learning strategies.

In one study, authors interviewed SafeCare providers in 11 sites to learn what adaptations their programs were making to better serve diverse families, and what additional adaptations they felt were needed. Authors found that, overall, providers perceived the model to work well with diverse populations and did not recommend systematically adapting the SafeCare intervention for specific populations. Instead, providers recommended individualizing SafeCare for specific local populations or families, an approach they were already taking. The modifications providers made to individualize SafeCare for local populations and families point to lessons about (1) strategies for family engagement, (2) the accessibility of program materials, and (3) the need for flexibility in scheduling sessions.

  • Family engagement. Providers emphasized the importance of engaging families and gaining trust during initial visits. Some providers observed that matching families to home visitors based on language was critical for engagement, and that matching based on race or ethnicity could help improve engagement. For families with extensive histories with child protective services, providers suggested explaining mandatory reporting requirements to build open communication and trust.
  • Accessibility of materials. Providers recommended making the materials more accessible by reducing the literacy levels, incorporating more pictures, adding culturally specific examples, and improving translations. Authors noted that lowering literacy levels and adding pictures could be valuable to SafeCare participants of all backgrounds.
  • Service flexibility. Providers recommended being open to meeting outside of families’ homes and being respectful of families who missed appointments for ethnic celebrations or rituals.

In a study of a statewide effort to integrate SafeCare into the state child welfare system, 295 individual providers of direct child welfare services from 50 private agencies were trained to deliver the program. In that state, public child welfare workers conducted child maltreatment investigations and then typically referred cases to direct child welfare service providers at private agencies. Researchers found that although providers were able to perform well during training and implement SafeCare in the field with high fidelity, the level of implementation was low: only 25 percent of providers implemented SafeCare following training. Researchers noted that even with free training and support, broad-scale implementation could not be achieved without planning at the organizational and systems levels. Specifically, providers noted two problems: (1) local child welfare staff were less aware of SafeCare than other existing programs and (2) SafeCare was added as a distinct service requiring referral (rather than being integrated into an existing program) and therefore competed for referrals with existing programs. The study authors noted that sufficient awareness about SafeCare was more difficult to achieve in larger states with many counties and agencies.

One study of CPAT’s development and initial testing described lessons related to CPAT implementation. The study authors reported that mothers responded positively to CPAT, feeling that the number of messages and calls was “just right.” Of the various types of texts received, mothers were more likely to respond to texts asking a direct question rather than those offering a supportive comment or a PAT prompt. Parents reported especially liking messages that suggested family activities to do in the community; parents often reported attending events that they otherwise would not have.