Licensure of Sheltered-Care Facilities: Does It Assure Quality?

Abstract

In California, licensure was intended to assure a minimum level of quality in sheltered-care facilities for the mentally ill population. This longitudinal study relates characteristics of facilities, their residents, and communities to subsequent licensure and considers differences between licensed and unlicensed facilities at follow-up. Initial interviews were completed in 214 facilities in 1973 six months before the implementation of the California Residential Facilities Licensing Act. Follow-up interviews occurred in 1985. Results indicate that although licensure occurred with greater frequency among facilities serving the most disabled population, licensure neither predicts nor has as its apparent consequence the development of higher-quality facilities. An alternative approach to quality assurance is offered.

During the past 35 years, the decentralization of responsibility for services to the mentally ill population has been the major guideline of mental health policy. Hospital units were decentralized geographically, and state responsibility was decentralized to county departments and later to catchment areas within counties (Mechanic, 1989). Furthermore, responsibility for the care and maintenance of individuals was decentralized to the nursing home system and the sheltered-care facility system; both systems now have provided for the care of more patients than mental hospitals ever did (Goldman, Gattozzi, & Taube, 1981). Although coordination of decentralized organizations has always been a problem, neither coordination nor quality-control issues have been given much attention in the literature. State legislatures, however, have used facility licensure to assure quality of care in decentralized systems.

This article looks at the effort to license sheltered-care facilities as an attempt to achieve quality control in a decentralized system. Sheltered-care facilities—including board-and-care homes, family care homes, halfway houses, and psychosocial rehabilitation facilities—are now the placement of choice for seriously mentally ill individuals who need supervised care. Therefore, it is necessary to guarantee that such care is quality care. Two questions about quality of care were raised for study: (1) What characteristics distinguish facilities that become licensed from those that do not? (2) Although we cannot attribute direct causation to licensure, what are the consequent differences between licensed and unlicensed facilities?

In the summer of 1973, Segal and Aviram (1977, 1978) completed a survey of 214 sheltered-care facilities, a representative sample of all such facilities in California that served mentally ill adults ages 18 to 65. Also in 1973 the State of California passed a licensing law for such facilities, the California Residential Care Facilities Licensing Act. Because the law did not go into effect until January 1974, the data from the Segal and Aviram sample were gathered before licensure was a possibility. Between 1983 and 1985 the Mental Health and Social Welfare Research Group of the University of California, Berkeley, reinterviewed or otherwise determined the status of each facility in the original study.

The Calfornia law required licensure of all 24-hour residential-care facilities that were not licensed as hospitals and that provided supervised living arrangements to mentally ill adults. However, facilities could avoid the licensure requirement by disavowing their claim to providing a supervised setting for the disabled population and by representing themselves as boarding houses.

The authors looked at several quality-of-care indicators that could influence licensure, including operational characteristics of a facility, its resident population, and its community environment. The most important operational characteristics include a facility’s cost to clients and the character of its environment; the latter was thought to be the main reason for licensing a facility. We assumed that the higher-priced facilities would be more likely to become licensed, because they could afford to comply with regulations that favored the development of intrafacility programming and that often required expensive physical plant modifications. We further hypothesized that the facilities would become more institutional in character, moving away from a traditional family atmosphere to a more structured social environment, because the structured environment would make it easier to comply with licensing laws. We also believed that the professional orientation of treatment facilities would make licensure more attractive to owners and managers.

In looking at the resident population of a facility, we expected that those facilities serving an increasingly disabled sample and those likely to serve mentally ill people as their major target group would be more likely to become licensed. Facilities who take everyone have a broader range of disability levels among their residents and greater access to different referrals, thus limiting their incentive to obtain a license.

In considering community environment, we hypothesized that neighbor complaints would draw attention to facilities and lead to licensure. Furthermore, facilities located at a distance from community resources would seek licensure to compensate for their peripheral status in the service system and, thus, to ensure continued flow of resident referrals. We also hypothesized that the physical environment of the facilities that became licensed would be viewed more positively by both interviewers and residents.

In examining the consequences of licensing, we again looked at a facility’s operational character, its resident population, and its community environment. In 1985 we asked a broader range of questions about operational character, including questions about tolerance of activities most people are able to do (“normal behavior”) and of activities that might pose problems in group-living situations (“deviant behavior”). We believed that licensed facilities would be less tolerant of certain normal and deviant behaviors because the facilities need to conform to a public moral standard. Finally, we hypothesized that licensed facilities would increase their use of health, mental health, and social services because licensing and becoming part of the formal system would make such services more readily available to them.

Methods

Sample and Data

Data in this analysis came from a longitudinal study of sheltered-care facilities. The 1973 sampling frame was based on state fire clearance records and state and county mental health department records. The facilities in the sampling frame included all family care, board-and-care homes, and halfway houses in California that served at least one resident who had a history of inpatient psychiatric hospitalization. These facilities were not licensed hospitals but did provide the minimum supervision necessary to enable residents to qualify for the protected-living arrangement rate under the Aid to the Totally Disabled Program (now the Supplemental Security Income [SSI] program). All of the 214 facilities in the original study were contacted again in 1985; 156 (73 percent) were still open at follow-up. We were able to interview the current manager in 151 of these.

Facility data were obtained from three sources: (1) face-to-face structured interviews with facility operators (on-site managers), (2) the 1975 to 1985 State of California Health and Welfare Agency Residential Care Facility Licensing Directories (the facility’s appearance in the state directory was taken as confirmation of licensure), and (3) assessments of the quality of each facility by the interviewers (social workers with at least one year of experience with the chronically mentally ill population).

Analyses

The dependent variable for the analysis of factors associated with subsequent facility licensure is a dichotomous indicator of whether the facility was licensed by the state during the follow-up period. The licensing status of all 214 facilities was investigated: 152 were licensed, 51 were not licensed, and 11 had a questionable licensing history. Managers of the latter 11 facilities claimed to be licensed; however, we were unable to confirm this claim in the state directories and therefore have deleted this group from the analysis. All independent variables in the analysis of subsequent licensure were grouped into three categories: facility characteristics, resident characteristics, and community characteristics.

Two types of statistical analyses are used: (1) bivariate comparisons of individual variables distinguishing features of licensed facilities from unlicensed facilities and (2) bivariate comparisons within facilities that described their residents as highly disabled and those facilities with a less-disabled group. This analysis follows the suggestion of Van Putten and Spar (1979) that poor facility environments may reflect more the extent of disability among residents than the intent of the facility operators.

The authors offer three methodological caveats. First, ideally, all characteristics related to subsequent licensure and those distinguishing between licensed and unlicensed facilities at follow-up would be the same measures. However, researchers upgraded the detail and quality of information in their follow-up interview schedules. Therefore, differences between licensed and unlicensed facilities in 1985 are reported in greater detail than in the 1973 data. Second, complaints received by facilities from neighbors during the follow-up period were not dated, and therefore we were unable to determine whether the complaints preceded or followed licensure. (Complaints reported in 1973, however, were clearly received prior to licensure.) Third, we did not randomly assign facilities to become licensed or remain unlicensed. We cannot directly attribute causation for observed differences between facilities at follow-up to licensure. Therefore, in this article we speak about apparent instead of actual consequences of licensure.

Measurement of Factors Associated with Subsequent Licensure

Facility Characteristics.

Program focus was measured by operator responses to Moos’s (1974, 1975) Community Oriented Program Environment Scales (COPES). COPES assesses three characteristics of the facility’s program environment: (1) relationship (including program involvement, support, and spontaneity); (2) treatment (including autonomy, practical orientation, personal orientation, and anger and aggression); and (3) system maintenance (including program clarity, order and organization, and staff control). Service availability and use were assessed by asking operators about 11 health, mental health, and social services. Separate scale scores were computed so that higher scores indicated greater availability and greater use of the services.

Additional quality assessment questions addressed to social worker interviewers inquired as to their relative willingness to move into the facility if their circumstances required a move to sheltered care and their overall assessment of the quality of the facility. The percentage of interviewers responding positively to each question is reported.

Resident Characteristics.

A scale was constructed to measure the degree of disability of the residents in the facility. Managers were asked how many residents were unable to be employed in the regular labor market, could never hold regular employment, were abandoned by their families, could not be self-supporting, and needed help with their basic life functions. Managers reported disability scores for each item on a six-point scale in which 1 = none, 2 = few, 3 = less than half, 4 = half, 5 = most, and 6 = all. The disability measure was a sum of the responses to the five items. Higher scores indicated that the facility served a more-disabled population.

Community Characteristics.

Operators were asked how far the facility was located from 11 community resources: (1) shopping center or local shopping area; (2) park; (3) library; (4) movie theater; (5) community center; (6) public school, high school, or college; (7) restaurant or coffee shop; (8) bar; (9) place of worship; (10) volunteer organization; and (11) barber or beauty shop. A scale describing relative distance to these community resources was created. Managers indicated how far it was to each of the 11 resources: 1 = definitely needed transportation, 2 = normally used transportation, 3 = a long walk from the facility, 4 = within easy walking distance of the facility, 5 = within one or two blocks of the facility. To calculate a distance score, scores were summed for each community resource and the total was divided by the number of resources evaluated. “Neighbor complaints” measured whether a facility had received at least one complaint during the study period.

Measurement of the Differences between Licensed and Unlicensed Facilities at Follow-up

Additional measures of facility characteristics used in the analysis of the apparent consequences of licensure include detailed cost assessments and interviewer ratings of the facility’s physical attractiveness, environmental diversity, and resident and staff functioning. The latter assessments are based on the Multiphasic Environmental Assessment Procedure (MEAP) (Lemke & Moos, 1986; Moos, Lemke, Mehren, & Gauvain, 1979). Sections of the MEAP reporting the facility’s tolerance for deviant and normal behavior were also considered.

Results

Predictors of Licensure

Facilities having a more system maintenance-oriented program and charging higher fees were significantly more likely to become licensed (Table 1). Facilities with more disabled residents were also significantly more likely to be licensed. Analyses of the predictors of licensure conducted separately for facilities with high and low numbers of disabled residents showed no additional significant differences.

Table 1.

1973 Characteristics Associated with Subsequent Licensure (N = 203)

Characteristic	Licensed Facilities (n = 152)	Unlicensed Facilities (n = 51)	t	df	p
Facility characteristics
Facility size (mean number of beds)	34.0	32.2			NS
Program focus (mean scores)
Relationship	11.49	11.93			NS
Treatment	10.25	10.38			NS
System maintenance	9.30	8.44	1.70	141	.045a
Availability of health, mental health, and social services (mean scores)	34.4	34.6			NS
Use of health, mental health, and social services (mean scores)	29.1	29.4			NS
Monthly charge per resident (mean dollars)	211	186	4.33b	193	.001
Interviewer’s positive evaluation of facility quality
Reaction to moving in as resident (%)	67.3	64.6			NS
Quality of the facility (%)	81.9	75.0			NS
Resident characteristics
Disabled residents (mean number)	18.5	17.0	2.42	198	.02
Mentally ill residents as the primary group (%)	67	61			NS
Community characteristics
Distance to community resources (mean scores)	3.6	3.6			NS
Neighbor complaints (%)	25.7	25.5			NS

^aBased on a one-tailed test.

^bBased on separate variance.

Licensure and Facility Closure

Of the 152 facilities that received a license during the follow-up period, only 15.8 percent closed. Of the 51 facilities that did not receive a license during the follow-up period, 66.1 percent closed (χ² [1] = 48.43, p < .000). It must be noted that when other facility characteristics, primarily financial investment in the facility business, are considered, licensure was still important in determining whether a facility remained open (Segal & Silverman, 1993).

Apparent Consequences of Licensure

In 1985 licensed facilities were more likely to use health, mental health, and social services than were unlicensed ones (Table 2).

Table 2.

Mean Differences between Licensed and Unlicensed Facilities in 1985 (N = 145)

Characteristic	Licensed Facilities (n = 128)	Unlicensed Facilities (n = 17)	t	df	p
Facility characteristics
Facility size (mean number of beds)	38.0	51.9			NS
Program focus (mean scores)
Relationship	11.32	12.25			NS
Treatment	11.6	10.4			NS
System maintenance	9.4	8.8			NS
MEAP (mean scores)
Physical attractiveness	.59	.58			NS
Environmental diversity	.57	.59			NS
Resident functioning	.55	.59			NS
Staff functioning	.70	.67			NS
Availability of health, mental health, and social services (mean scores)	35.1	34.0			NS
Use of health, mental health, and social services (mean scores)	31.2	27.3	1.81	139	.04a
Monthly cost (mean dollars)
Total monthly expenses	11,735	3,048	4.67b	53	.001
Mortgage or rent	2,269	1,040	2.3lb	21	.03
Utilities	1,028	880			NS
Transportation cost	379	130	3.32b	20	.005
Program cost	2,519	113	3.75b	89	.001
Food	2,644	1,593			NS
Other expenses	3,028	642	3.14b	77	.003
Monthly cash flow (monthly income – expenses; mean dollars)	4,361	283	4.25b	83	.001
Monthly charge per resident (mean dollars)	513	347	4.24b	39	.001
Resident characteristics
Disabled residents (mean disability scale scores)	20.6	18.3	1.8	127	.04a
Mentally ill residents as the primary group (%)	63	71			NS
Community characteristics
Distance to community resources (mean scores)	2.9	3.7	−1.97	125	.05
Neighbor complaints (%)	28.3	3.9	13.14	1	.001

Note: MEAP = Multiphasic Environmental Assessment Procedure (Moos, Lemke, Mehren, & Gauvain, 1979; Lemke & Moos, 1986).

^aBased on a one-tailed test.

^bBased on separate variance estimate.

At follow-up the licensing status of a facility was significantly related to the cost of running the facility. Licensed facilities had higher monthly expenses than unlicensed ones. Licensed facilities spent significantly more in all categories of monthly expenses (for example, rent, transportation, program cost) except utilities and food, had higher cash flows (monthly income minus monthly expenses), and had a higher monthly charge per resident.

Among the characteristics of residents, only resident disability scale scores distinguished licensed and unlicensed facilities. Licensed facilities showed a tendency to have more severely disabled residents.

In considering community characteristics, licensure was significantly associated with neighbor complaints. During the follow-up period, licensed facilities were more likely to receive a complaint than were unlicensed ones. However, unlicensed facilities were significantly closer to community resources. For unlicensed facilities the average distance to community resources reported was between “within easy walking distance of the facility” and a “long walk from the facility,” whereas that of licensed ones approaches a “long walk.” The closeness of unlicensed facilities to community resources may be one reason they are able to remain open and avoid licensure; such facilities have easy access to referrals.

Finally, we considered the 1985 assessments of facility tolerance for certain normal and deviant behaviors. Total scale scores indicated that licensed facilities were significantly less likely to tolerate normal behaviors (Table 3). The differences derive largely from item differences regarding taboo behaviors. Licensed facilities were less likely to tolerate drinking and sexual behaviors than were unlicensed settings, suggesting that licensed facilities are more institutionalized than unlicensed ones. These results do not change when controlling for the degree of disability of residents living in the facility.

Table 3.

Differences in Tolerance for Normal and Deviant Behavior in 1985

	% of Managers Who Tolerated
Behavior	Licensed Facilities	Unlicensed Facilities
Normal
Drinking liquor in one’s own room	4.7	18.8*
Having one’s own furniture in the room	80.5	93.3
Moving furniture around in the room	73.4	87.5
Keeping a fish or bird in the room	57.0	62.5
Keeping a hot plate or coffee maker in the room	15.7	20.0
Doing some laundry in the bathroom	62.5	73.3
Drinking a glass of wine or beer at meals	18.8	37.5**
Skipping breakfast to sleep late	43.0	56.3
Closing the door to one’s own room	98.4	100.0
Locking the door to one’s own room	38.1	56.3
Having sexual relations with a friend at the house	29.1	50.0**
Deviant
Refusing to participate in programmed activities	50.8	53.8
Refusing to take prescribed medicine	14.1	6.7
Smoking in bed	1.6	12.5*
Being drunk	1.6	0.0
Leaving the building during the evening without letting anyone know	10.9	56.3*
Refusing to bathe or clean oneself regularly	0.8	6.7**
Creating a disturbance or being noisy	4.7	6.3
Stealing others’ belongings	0.8	0.0

^*p < .05.

^**p < .10.

Although no significant differences in total scale scores were found in the deviant behavior scale, licensed and unlicensed facilities differed significantly in three of the eight individual scale items (Table 3). Compared to unlicensed facilities, licensed facilities were less willing to tolerate residents smoking in bed, leaving the building without notice, and refusing to clean themselves regularly.

Discussion

The positive aspects of licensure seem largely confined to its consequences rather than its influence of selecting facilities into this system of public accreditation. Following licensure facilities began using more health, mental health, and social services, which may have resulted from the formalization of the relationship between the facility’s management and the mental health system and enhanced their chances of survival.

Licensure also seemed to increase public visibility. Citizens can express their concerns regarding care offered to facility residents through state licensing departments. Thus, at follow-up we observed significantly more complaints made about licensed than unlicensed facilities. (Although we cannot be assured that these complaints occurred following licensure, the fact that complaints before 1973 did not predict licensure supports this assumption.)

Licensure’s influence on facilities has both positive and negative aspects. That facilities whose environment is more focused on system maintenance tended to become licensed raises concern regarding increased risks of institutionalization in environments that are supposed to be antidotes to this problem. This concern is further heightened by the differences between licensed and unlicensed facilities in their tolerance of socially taboo yet often normal behavior. Differences between licensed and unlicensed facilities in their tolerance for smoking in bed and lack of bathing clearly indicate the dilemma faced by advocating normal environments within facilities serving this population. Perhaps an educational campaign directed at facility operators regarding the importance of normal behavior and the need to set appropriate limits on smoking in bed and hygiene would help reduce the significance of this dilemma.

The facilities that served the most disabled people and that maintained more disabled people also tended to become licensed. The benefit is that public attention is focused on protecting the care of the most disabled population. The drawback is that disability was unrelated to either cost or quality in distinguishing between licensed and unlicensed facilities. Licensed facilities always cost more regardless of the level of disability of their residents. One might speculate that licensure attracts only those facilities that can afford it and becomes a justification for perpetuating cost differentials regardless of quality.

The lack of significance of quality indicators raises questions about the use of licensure to assure quality of care. Multiple indicators of quality of care were used in our assessments, including direct evaluations by interviewers, assessments of the availability and use of services, descriptions of program focus, assessments of the physical attractiveness and environmental diversity of the facility, and resident and staff functioning. With the exception of use of health, mental health, and social services in 1985, no indicators significantly distinguished between licensed and unlicensed facilities.

In 1985 we were able to break down costs into the facilities’ operational budget components. Quality-of-care advocates for licensure have frequently cited rumors regarding poor facilities and cuts in food costs. Whereas significant differences between licensed and unlicensed facilities were obtained for most expenditure indicators, no differences were observed in food costs.

Recent state budget crises experienced throughout the United States heighten the importance of these findings, because licensing departments have been chronically understaffed and are only able to address the most serious facility violations. Therefore, it may be unfair to conclude that licensing cannot assure quality. We can say that under current conditions, California’s strict, detailed, quality assessment-driven licensing law seems unrelated to selecting or distinguishing quality facilities.

What then is the function of licensing if not quality-of-care assessment? Perhaps the best function of licensing should be limited to policing abuse of residents and the care of the facility’s physical plant, thus contributing to the development of certification organizations whose function would be to highlight the positive aspects of facilities. We recommend the development of a guide that recommends quality residential care facilities to the public. Such a guide could also enable licensing agencies to focus their limited resources on the most serious violations of resident rights and facility physical plant problems. Some licensing organizations have already been forced to use this strategy. Recognizing their limited capabilities will allow more informal evaluative efforts to be legitimized. Attempts to develop certification procedures could be endorsed and promoted by state departments of mental health without overlapping or conflicting with licensing efforts.

Conclusion

Efforts to decentralize responsibility for the care and treatment of mentally ill people that depend on licensing for quality control need to be seriously reconsidered. Licensure is apparently unrelated to bringing quality facilities into the system or to maintaining quality of care in facilities. Licensure does focus public attention and accreditation on facilities serving the most disabled people. Unfortunately, licensed facilities always cost more.

Licensing seems essential to protecting against major violations of resident rights and physical plant codes. We recommend the development of certification procedures that emphasize the positive aspects of sheltered care to focus more attention on the provision of quality care. ■