11.4.1. Outpatient care

Table 185

Summary clinical evidence profile: Comparison 2. Cohort segregation by location versus no cohort segregation.

Table 186

Summary clinical evidence profile: Comparison 3. Combination of protective equipment + individual segregation versus incomplete protective equipment + incomplete individual segregation.

11.4.2. Inpatient care

Table 187

Summary clinical evidence profile: Comparison 4. Cohort segregation by location versus no cohort segregation.

Table 188

Summary clinical evidence profile: Comparison 5. Individual segregation by location versus usual care.

11.4.3. Combined inpatient and outpatient care

Table 189

Summary clinical evidence profile: Comparison 6. Cohort segregation versus no cohort segregation.

Table 190

Summary clinical evidence profile: Comparison 7. Complete cohort segregation versus incomplete cohort segregation.

Table 191

Summary clinical evidence profile: Comparison 8. Individual segregation versus usual care.

Table 192

Summary clinical evidence profile: Comparison 8. Individual segregation versus usual care.

Table 193

Summary clinical evidence profile: Comparison 10. Cohort segregation + individual segregation + protective equipment versus usual care.

11.6.1. Outpatient care

11.6.2. Inpatient care

11.6.3. Combined inpatient and outpatient care

11.6.4. Economic evidence statements

No published evidence on cost-effectiveness in people with cystic fibrosis was available for this review.

The economic model found that cohort segregation according to chronic P aeruginosa was cost-effective compared to no cohort segregation (3 studies).

The economic model found that cohort segregation according to intermittent P aeruginosa was not cost-effective compared to no cohort segregation (2 studies).

The economic model found that cohort segregation according to intermittent B cepacia complex was cost-effective compared to no cohort segregation in 1 study and cost-ineffective in 1 study.

The economic model found that the addition of protective equipment was cost-effective to prevent cross-infection with intermittent B cepacia complex (1 study) and intermittent P aeruginosa (1 study).

The economic model found that individual inpatient segregation (single inpatient rooms) according to intermittent B cepacia complex would be cost-effective compared to no individual inpatient segregation (shared ward) (2 studies).

11.7.1. Relative value placed on the outcomes considered

The aim of this review was to determine the effectiveness of the different strategies, such as cohorting, segregation, or protective equipment, in reducing the transmission of cystic fibrosis pathogens.

The committee identified incidence and prevalence of patients infected with transmissible pathogens as critical outcomes for decision making. Quality of life, emotional function, patient satisfaction, staff experience and staff and patient compliance were rated as important outcomes.

11.7.2. Consideration of clinical benefits and harms

The committee noted that there were existing NICE guidelines on preventing and controlling infection. Therefore, they incorporated a reference to these guidelines and focussed their recommendations on cross-infection concerns specific to people with cystic fibrosis.

The committee noted that the evidence showed mixed results with regards to whether incidence and prevalence of people infected with transmissible pathogens decreased after cohort segregation was implemented across the inpatient and outpatient setting or in the outpatient setting alone. There was some evidence that the incidence of B cepacia complex and hospital-associated P cepacia infections decreased after cohort segregation was implemented in the inpatient setting alone. Moreover, there was some evidence that prevalence of B cepacia complex infection decreased after individual segregation was added to cohort segregation across the inpatient and outpatient setting. The committee noted the lack of evidence assessing individual segregation as an intervention on its own in the outpatient setting alone. The committee noted that feelings of isolation may be a potential disadvantage of individual or cohort segregation. However, there was some evidence that people with cystic fibrosis and carers supported both cohort and individual segregation.

The committee noted that infection control would have to be implemented both in the inpatient and outpatient setting at the same time. Therefore, the committee recommended that arrangements to prevent cross-infection among people with cystic fibrosis should be based on a local infection control strategy that covers both outpatient and inpatient care. This strategy would cover interventions such as cleaning of rooms and equipment, closing the doors of rooms in the hospital or the outpatient clinic and effective ventilation in gyms to remove exposure of airborne cross-infection. Moreover, the committee recommended that arrangements to prevent cross-infection should also be based on a microbiological surveillance programme.

The committee noted that segregation interventions would be ineffective if people did not adhere to segregation rules or they met with each other outside of the health care setting. Therefore, a recommendation was prioritised to inform people with cystic fibrosis, their family members or carers and staff involved in their care about the risk of cross-infection and how to avoid it.

The committee agreed that a combination of cohort segregation and individual segregation is likely to be more effective than cohort segregation alone. Cross-infection can occur between people with cystic fibrosis who are designated as having the same pathogen but have different strains which may have differing virulence characteristics. For example, superinfection with different, more virulent epidemic strains of P aeruginosa, potentially as a result of cross-infection, can occur in people who already have P aeruginosa. Similarly, patients with less virulent species of B cepacia complex, for example Burkholderia multivorans, could become infected with the more virulent B cenocepacia if cohorting is based on B cepacia complex status without species differentiation or if patients have as yet undetected infection. If people were to be cohorted based on different pathogen strains, cohort interventions would become complex due to a relatively high number of different cohorts which may increase as understanding of individual pathogen virulence develops. Moreover, any information on a person’s infection status refers to the last sputum culture that was performed. Therefore, after acquisition of an infection there would be a time interval until the next sputum culture. During this period someone would be cohorted with people who are free of infection. Therefore, the committee concluded that separating all people with cystic fibrosis, regardless of their infection status, from each other was very important both in the outpatient and inpatient setting.

The committee agreed that each specialist cystic fibrosis clinic should be organised as to ensure that contact between people with cystic fibrosis is prevented both during the use of communal areas such as waiting areas, cafes and restrooms, and during attendance at diagnostic, treatment and pharmacy facilities. The committee noted that there were multiple options to separate people; for example, placing the person directly into a room, where the multidisciplinary team can go to conduct the visit.

The committee agreed that even if a clinic separates all people from each other, casual contact may still occur. Therefore, the committee agreed that a combination of cohorting and individual segregation was more effective than individual segregation alone and recommended that the local infection control strategy that covers outpatient and inpatient care should include cohorting. The committee noted casual contact might be more likely to happen in the outpatient setting because the whole patient journey would need to be taken into account. Additionally, it is more common that people attend for the use of diagnostic or treatment facilities without a prior booking in the outpatient setting.

Therefore, in addition to individual segregation, the committee agreed to prioritise a recommendation to keep people with transmissible or chronic P aeruginosa or B cepacia complex infection separate from people who do not have these infections, for example by using separate outpatient clinics. The committee noted that one way of cohorting would be to have separate buildings for separate patients, another way of cohorting would be to book appointments with people with a specific pathogen on one day of the week and for people with a different pathogen on a different day of the week.

Intermittent P aeruginosa infection is harder to diagnose given that tests with positive results would alternate with tests with negative results. Therefore, the committee noted that it would be difficult to cohort people with intermittent P aeruginosa infection from those who do not have this infection. Moreover, some evidence showed a higher prevalence of intermittent P aeruginosa infection after cohort segregation in the outpatient setting. However a lower prevalence of chronic P aeruginosa infection was observed in the same setting. Therefore, these trends may be due to treatment strategies rather than to cohort segregation because treatment may prevent the intermittent pathogen from becoming chronic. With regards to this point, the committee agreed that changes in prevalence and incidence may be due, not only to cross-infection, but also to other factors such as acquisition of infection from environmental sources and advances in treatment. Moreover, prevalence relates to survival. The evidence also showed no clinically significant difference in the incidence of intermittent P aeruginosa after cohort segregation across the inpatient and outpatient setting. Given the lack of evidence supporting segregation and considering the difficulties involved in diagnosing intermittent P aeruginosa, the committee decided not to be overly prescriptive and recommended to consider keeping people who have intermittent isolation of P aeruginosa separate from people who do not have this infection, for example by using separate outpatient clinics.

The committee noted that no evidence was identified on non-tuberculous mycobacteria. Therefore, no recommendation was drafted specific to this pathogen. However, the committee noted that all recommendations on cross-infection except for those two that refer to P aeruginosa and B cepacia complex should apply to all people with cystic fibrosis, irrespective of their pathogen status. Cohorting could be done based on any pathogen that a local cross-infection control strategy may deem relevant.

The committee noted that the inpatient setting is where people would be more likely to come into close contact with each other as they may spend days or weeks in the hospital. Therefore, the committee prioritised a recommendation to give people with cystic fibrosis individual rooms with en-suite facilities. This recommendation is consistent with the NHS service specifications for cystic fibrosis. The committee noted that, in practice, people with cystic fibrosis are sometimes segregated into individual rooms but have contact in communal areas, this is to be avoided. Therefore, the committee recommended to help inpatients with cystic fibrosis plan their attendance to avoid contact with each other, for example when they use hospital restaurants, schools, recreational areas (such as the gym) and diagnostic, treatment and pharmacy facilities. The committee noted that a timetable could be used for school and gym attendance. Moreover, for diagnostic and treatment procedures there could be a computerised system suggesting who should enter specific areas at what time. Finally, systems should be in place to ensure that if someone with cystic fibrosis comes to the hospital, contact with the inpatients is prevented.

There was some evidence that prevalence of people infected with transmissible pathogens decreased after a combination of individual segregation with protective equipment was implemented in the outpatient setting alone. Incidence of infection decreased after a combination of cohort and individual segregation and protective equipment was implemented across the inpatient and outpatient settings. However, there was no evidence on the use of protective equipment alone. The committee noted that the use of protective equipment is difficult to implement. People’s motivation is limited because of stigma and because it protects other people but not themselves, as a result they tend to take it off. Therefore, the committee decided not to make a recommendation on protective equipment.

11.7.3. Consideration of economic benefits and harms

11.7.4. Quality of evidence

All the evidence included in the review was of very low quality as assessed by GRADE.

One RCT was included. This study had unclear risk of selection bias, performance bias, detection bias, attrition bias and outcome reporting bias.

Most included studies were retrospective uncontrolled before-after studies. In some of these studies, the intervention and comparison groups were drawn from years, or time intervals, that were considerably distant in time. These time intervals increased the likelihood that the intervention group may have been exposed to additional interventions affecting incidence or prevalence. Moreover, in some studies the frequency of cultures performed to detect relevant infections was unclear or only annual. Therefore, people that were considered “at risk” when the intervention started may have already had the infection. Consequently, there was a high risk of selection bias. Most studies did not control for any factor, therefore, there was a high risk of comparability bias. Furthermore, many studies looked at changes in prevalence and incidence without making a distinction between transmissible strains and unique strains. Most studies did not use genotyping in order to understand whether changes in incidence or prevalence were related to cross-infection. Therefore, there was high risk of bias in relation to the outcomes of incidence and prevalence.

Three surveys were included. Only 1 study assessed comparability between respondents and non-respondents and all 3 studies had high risk of selection bias. None of the 3 studies controlled the analysis for possible confounders, therefore, there was high risk of comparability bias.

11.7.5. Other considerations

There was some evidence that staff adherence to cohort segregation was high. This indicated that cohort segregation was feasible.

No equality issues were identified by the committee for this review question.

The committee discussed the need for a research recommendation in this area, but they agreed it was not needed. They noted it is the responsibility of local areas to devise their own policies and is primarily a public health issue. They acknowledged there is a NICE guideline on infection.

11.7.6. Key conclusions

The committee concluded that people with cystic fibrosis should avoid contact with each other irrespective of their infection status. Moreover, cohorting should be implemented in addition to individual segregation, to reduce chances of casual contact between people with different infection status. The minimum requirement in terms of cohorting is that people with transmissible or chronic P aeruginosa or B cepacia complex infection should be kept separate from people who do not have these infections, for example by using separate outpatient clinics. Additionally, outpatient and inpatient care facilities should follow their local infection control strategy to prevent cross-infection.