Evidence Behind the Green House and Similar Models of Nursing Home Care

Sheryl Zimmerman; Lauren W Cohen


Aging Health. 2010;6(6):717-737. 

In This Article


It must first be acknowledged that the limitation of this article in informing The Green House and similar models of NH care is that by deconstructing the elements as has been done, the strengths and limitations of each element in isolation may not reflect their combined effect – which is, after all, the way they are expressed in practice. However, some of the literature clarified the relationship between these elements, such as that choice and control relate to increased engagement, especially for residents with cognitive capacity.[15,135,137] Therefore, some elements may depend on others: some may be synergistic, and some may cancel out one another, in the case of more impaired resident case mix making consistent assignment less attractive. However, it must also be recognized that the literature is not quite as deconstructed as first appears, since none of the elements exist in a vacuum. For example, small NHs tend to have different care processes. In most cases, it is not possible to determine with certainty the separate effects of each element.

There are also important caveats to the findings reported in the literature. In the majority of instances, they report associations based on cross-sectional data; while a strength of these studies is that they allow examination of a large number of NHs and residents and may provide robust results, they cannot be used to infer causal relationships. Hence, the finding that consistent assignment relates to lower resident quality of life,[80] for example, can no more be assumed to suggest that assignment affects resident quality of life than it can be assumed to suggest that quality of life affects staffing patterns. Furthermore, there may be other matters that influence both staffing and quality of life, such as case mix. Another consideration of these large data sets is that small differences that may not be clinically significant may achieve statistical significance. In addition, even in large datasets, but especially in smaller datasets, there may not be a sufficient number of cases to detect important relationships that exist. Additional considerations important to intervention projects are that by relying on small samples or not using random assignment, detected differences may not be replicated. Hence, the conclusions to be drawn in this discussion are based on the preponderance of the data, because consistent results are likely more reliable and valid.

There is a final consideration when interpreting these findings in the context of The Green House homes, and it is that there is variability within these homes themselves. For example, one home had a resident eat meals at the kitchen counter instead of at the table because he was easily distracted.[3] In terms of staffing, while all homes have nursing staff nearby on the campus, the manner in which they work with the shahbazim may reflect different models.[147] This variability is important, because it is necessary that an innovation be able to be refined to meet specific needs if it is to be widely adopted.[11] Thus, it is with an appreciation of the necessary variation within The Green House homes that this discussion precedes.

Physical Structure: Small Size, Private Rooms & Bathrooms, & Access to the Outdoors

While all Green House homes have 12 or fewer beds, studies that examined the association of size with care and outcomes did not do so in relation to this size cut-off point. Furthermore, because few or no such size NHs were included in these studies, it is not possible to determine whether results would have differed in relation to homes with 12 or fewer beds. Thus, it cannot be determined from the literature whether having 12 or fewer beds compared with another size is the optimal cut-point. That said, small NH size favors psychosocial care, behavioral outcomes and some medical outcomes, while larger size conveys more resources, financial stability and ADL care. The psychosocial outcomes include better rapport, less abuse, more social involvement for impaired residents, and fewer conflicts and emotional and behavioral symptoms, presumably because the staff know the residents better and there are fewer environmental demands for impaired residents. The detriments to small size are the diseconomy of scale and fewer resources. For example, larger NHs may be more likely to achieve operational efficiencies and invest in augmented services and targeted quality improvement.[26] However, because The Green House homes are often part of a larger campus or grouped with other such homes, it is possible that this challenge may be overcome. It has been suggested that at least two Green House homes are needed to achieve economies of scale in staffing, and even more may be optimal.[206] This is an area for future research. Of note, both smaller and larger NHs conveyed benefits to medical and functional outcomes, suggesting that both closer staffing and resources are beneficial in this regard.

The literature related to privacy strongly supports the benefits of private rooms and bathrooms, especially for infection control and resident and family preference (although it is not considered the most important NH feature, being rated as less important than safety, medical care, food and staff).[202] Administratively, private rooms and bathrooms are advantageous in that they are easier to market and clean, and time is not needed to address roommate conflict and room transfers. However, the isolation can be problematic for some and result in less social engagement. Thus, the need to attend to social engagement may be especially important in The Green House homes, and more so for residents without dementia who tend to spend most of their time alone in their room.[3]

The literature regarding access to the outdoors is uniformly positive in relation to overall enjoyment and the benefit of light to regulate the circadian cycle. A study of four Green House homes found that residents with dementia used the space frequently, sometimes with assistance and sometimes independently. However, there were some complaints, including that the area was isolated, needed more sun protection and was difficult to access.[3] Similarly, research in traditional NHs suggests that outdoor space is underused, and more attention is needed regarding access, safety and sun protection.

Dining (Noninstitutional, Residential-style Kitchen)

The literature strongly supports the benefit of The Green House model of communal eating, primarily to benefit food intake. It also allows for staff assistance during eating, and in that regard it is informative that in both traditional NHs and The Green House homes, the resident:staff ratio during mealtimes is approximately 5:1.[62] In two of the first Green House homes specializing in dementia care, if clinical staff or visitors were present at the time, they provided eating assistance.[3] Of note, a more impaired resident case mix could challenge the staff, especially when others are not available for support.

The structure of the dining area in The Green House homes is such that there is one dining area, and all residents, staff and visitors eat around a common table. However, aggressive behaviors are most likely to occur in resident rooms and dining rooms,[40] and so the merits of this arrangement are important to consider. The postoccupancy evaluation of the first Green House homes found some residents and family did not like the noise, congestion and chaos resulting from the large dining table, and in at least one instance, a resident ate at a nearby counter to reduce distraction.[3] One study conducted in a traditional NH that promoted homelike ambiance and used more than one table demonstrated that intake of food was improved,[61] and so it is possible that an alternate seating configuration may be worthy of consideration.

The Green House homes also intend to use the house kitchen as a hub for activity and promoting a sense of community. While the postoccupancy evaluation of the first four Green House homes found one or more more residents with dementia were often seated at the table at nonmeal times, and in the other homes visitors sometimes congregated there, residents who were not cognitively impaired spent more of their time in their rooms.[3] Thus, there is indication that merely having this space is not sufficient to promote a sense of community, a point that will be further addressed in this article.

Finally, while the literature related to traditional NHs does not inform this point, it is important to note that the postoccupancy evaluation found the food actually served was not as personalized as intended in The Green House model. In some cases, menus were identical across homes and became dictated by the food service contractor, resulting in food that appeared less home-like.[206] Not only is this contrary to the model, but it is also contrary to that recommended by the American Dietetic Association, which suggests a diet in which residents have control over food choice.[148] Thus, adherence to the model may require diligence in regard to personalized meals.

Staffing (Consistent Assignment, Self-managed Teams & Clinical Staffing)

It is the area of staffing where the literature suggests careful attention to the elements of The Green House model as envisioned. Consistent staff assignment and self-managed teams have an inherent logic, but the literature is not strongly supportive of either one. It seems important to make this point more generally known, because consistent assignment has become a cornerstone of culture change, promoted by the Advancing Excellence campaign and embraced by a third of participating providers.[207] Data indicate that while residents and families prefer consistent assignment, staff preference is variable, it does not relate to staff turnover, and it may negatively relate to care and absenteeism. Of note, one study found consistent assignment matched a primary CNA to an individual resident only 50% of the time,[76] and so some of these data may not provide a valid reflection of what is presumed to be conveyed by 'consistent assignment'. At present, consistent assignment is best considered a preferred but not unqualified best practice.[78] Thus, it might be worth considering a rotating model of consistent assignment, which allows sufficient consistency to promote familiarity but also allows staff to better share their workload and come to know more residents. The Green House shahbazim may be especially able to share their workload given that they function as a team.

Similarly, some challenges have been detected with the use of self-managed work teams. While little research has been conducted of such teams in isolation of other elements of care, either no or modest positive findings were achieved related to nurses' and families' perceptions and CNA attitudes related to scheduling.[82] The most notable challenge to such teams seems to be their administration – the time needed to meet as a team and potentially insufficient leadership skills to manage the team effectively. Thus, there is an indication to conduct meetings more flexibly, and also to assure that the staff have the training, skills and personality to self-manage, which may not always be realistic. Successes, but also challenges, to self-managed work teams are borne out in the experiences of The Green House homes, where it was concluded that without sufficient training there was little sense of teamwork or perception that the team could be helpful in problem solving.[206] On the other hand, staff in The Green House homes reported better teamwork than those in comparison homes (perhaps due to a smaller staff pool) but also challenges related to interactions.[208] Of note, there is a suggestion that more supervision may be needed when residents are more impaired,[87] and in some cases not all decision-making may stay within the team.[89] A supportive supervisor may be especially important if self-managed teams are to be maximally effective.

Finally, The Green House homes are unique in that clinical staff are not based on The Green House unit, although they are usually nearby on the campus. The only data to compare to this model relate to physicians, who typically are not housed within a NH; in this regard, data uniformly indicate that increased presence is preferable. In terms of actual staffing hours, increased nursing, other clinical staff, administrative and CNA time relate to beneficial care and resident outcomes in areas including pressure ulcer incidence, infections, mortality, abuse and engagement. The Green House model does not promote less clinical staffing, and indeed the workflow study indicates that these homes had 1.6 more HPRD of shahbazim (i.e., CNA) time than traditional NHs (including up to 30 min more of direct care) and a similar amount of RN/LPN time. However, 1.8 of shahbazim 4.2 HPRD were spent on indirect care, such as meal preparation, housekeeping and laundry.[6,208] Furthermore, it must be considered that the skill mix of staff is important, and the disproportion of direct care staff to RNs and LPNs may be challenging when caring for more impaired residents. This is clearly an area for research, especially with regard to resident outcomes. In addition, it has been raised as a concern by families of The Green House residents, some whom felt that having two direct care staff was insufficient in cases of emergencies, and some who were concerned about not having nurses in the building.[5]

Of note, variability has been evidenced in the manner in which nursing staff work with the shahbazim, in that it follows either a traditional hierarchical nursing model (which was disappointing to some shahbazim), a divided nursing model (in which the nurses did not work as a team with shahbazim), an integrated nursing model (which related to feelings of teamwork but also more work for the nurses as they spent time in direct care) or a visiting consultant model (which resulted in nurse dissatisfaction, less nurse oversight and missed indicators of change in residents' condition).[147] In addition, nurses in The Green House homes reported more stress related to communicating with the care team than those in traditional NHs.[208] Thus, it seems necessary to be more prescriptive as this collaborative model develops. Regarding the other clinical staff, occupational and physical therapists seemed to adapt to providing care in the home; the medical director noted that it was more complex to make medical rounds there than in traditional NHs, and the activity director and social worker did not change their services.[206] As addressed in the Engagement section later in this article, there is cause to reconsider the role of the activities personnel. Furthermore, it appears that the role of the social worker has not adapted itself to this new model and is not being maximally beneficial.

Elder Case Mix (Diverse & Stable)

The intent of The Green House homes is that there be equitable access and that their residents reflect those who live in traditional NHs. In addition, the need to allow Medicare-funded stays is important for the financial viability of these homes. In the study of 14 Green House homes compared with 13 traditional homes, The Green House residents were not as functionally impaired in dressing, transferring, toileting or eating, and fewer had behavioral problems, were on tube feeding or did not communicate; furthermore, payer sources for the Green House homes were 5% Medicare and 39% Medicaid,[6,208] which is lower than the national resident-level percentages of 13 and 60% shown in Table 1. Thus, if these Green House sites are typical of others, they have not yet achieved a case mix that is reflective of traditional NHs. As to the capacity, in the limited data that exist regarding postacute care, therapists indicate that The Green House homes are suitable for therapy provision except when very large equipment is needed.[206] However, it is not clear that the staffing arrangements of The Green House homes are suitable to care for postacute residents.

The intent underlying equitable access is important, but it is also important that residents in The Green House homes have enough similarities and time with each other and with staff to forge social bonds, as those relationships are considered the primary source of engagement and relate to health and well-being. Hence, it may be best that within any given home, resident homogenity be the goal, with heterogeniety achieved across homes. However, staffing may need to be readjusted because some homes may have residents with more care needs – such as will be the case if future sites are targeted to serve short-stay residents or those with multiple sclerosis. In reference to social bonds, there are no strong data to inform the extent of similarity that is optimal, and this topic is worth exploring. Similarly, there are no data to inform the duration of time required to forge social bonds, and this surely is an individualized matter.

Elder-centered Care (Not According to Fixed Schedules)

The literature raised four compelling findings related to choice and control: that control is important; that the desire for control may not be universal nor beneficial; that predictability may be more important than control; and that control/predictability must be sustained. While control remains important as individuals age, they come to accept that more events are beyond their own control. Thus, relinquishing control may not be as distasteful to older adults as it is to younger individuals. Consistent with this point, NH staff do not understand the value that residents place on controlling different components of their daily life, as they undervalue the importance of control over the outside world and overvalue the importance of control over the inside world.[138] Findings that predictability is as important as control have important implications for care because it is markedly easier to assure consistency than it is to make standard routines highly individualized; furthermore, such personalization may leave less time to address other areas of care, and so what is offered for individualization must be realistic. Hence, it is beneficial to allow resident control, but elder-centered care must itself reflect the individual's preference to exercise control; in addition, the trade-off of allowing elders to maximally direct their activities must be balanced.

Engagement (Normalized Rather Than Organized Activities)

A consistent finding is that not only is social engagement beneficial for function and well-being, but it is often necessary to actively engage residents in social and recreational pursuits. Resident engagement in NHs is typically modest, with less than 10% participating in activities most of the time;[140] however, participation increases dramatically when materials and prompting are provided. Thus, while some disengagement may reflect resident choice, it may also reflect the need for staff to more actively promote engagement – especially for those with cognitive impairment and those who are functionally impaired. Furthermore, involvement in organized activities is higher when others participate, suggesting the beneficial nature of group activities. However, case mix is an important consideration, as residents who are less impaired participate in more activities when there is more autonomy, but this is not true of those who are less impaired.

In terms of The Green House homes, they provide fewer structured activities than do traditional NHs[6] – a concern noted by families[5] – and similarly, their residents are less likely to participate in organized activities, presumably due to the absence of dedicated, on-site activities staff.[4] The intent of The Green House homes is that residents will assist in normative activities, such as washing dishes, preparing food, folding the laundry and cleaning, and that these and informal interactions will substitute for formal activities. In reality, however, few residents participate in domestic activities, social activities do not arise automatically and shahbazim do not facilitate interaction.[3] Given the benefit of social engagement and the need to actively promote it – while recognizing that certain personality characteristics may be necessary to do so – it seems that a more active role for formal activity personnel, and more training and support of the shahbazim in this regard is indicated in The Green House homes.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: