Description of Quarantine in Toronto

During the first and second SARS outbreaks in Toronto, >15,000 persons with an epidemiologic exposure to SARS were instructed to remain in voluntary quarantine (Health Canada, unpub. data). Data on the demographics of the quarantined population were collected, but have not yet been analyzed (B. Henry, Toronto Public Health, pers. comm.). Quarantined persons were instructed not to leave their homes or have visitors. They were told to wash their hands frequently, to wear masks when in the same room as other household members, not to share personal items (e.g., towels, drinking cups, or cutlery), and to sleep in separate rooms. In addition, they were instructed to measure their temperature twice daily. If any symptoms of SARS developed, they were to call Toronto Public Health or Telehealth Ontario for instructions (5).

Study Population

All persons who were placed in quarantine during the SARS outbreaks in Toronto (at least 15,000 persons) were eligible for participation in this study. The survey was announced through media releases, including locally televised interviews with the principal investigators. Information on the study and invitations to participate were posted in local healthcare institutions, libraries, and supermarkets. Ethics approval was obtained from the research ethics board of the University Health Network, a teaching institution affiliated with the University of Toronto.

Survey Instrument

A Web-based survey composed of 152 multiple choice and short- answer questions was to be completed after participants ended their period of quarantine. It took approximately 20 minutes to complete. Questions explored included the following: 1) knowledge and understanding of the reasons for quarantine (2), knowledge of and adherence to infection control directives, and (3) source of this knowledge.

The psychological impact of quarantine was evaluated with validated scales, including the Impact of Event Scale—Revised (IES-R) (9) and the Center for Epidemiologic Studies—Depression Scale (CES-D) (10). The IES-R is a self-report measure designed to assess current subjective distress resulting from a traumatic life event and is composed of 22 items, each with a Likert rating scale from 0 to 4. The maximum score is 88. In a study of journalists working in war zones, the mean IES-R score of posttraumatic stress disorder (PTSD) was 20. In these persons, the presence of PTSD symptoms, as measured by this scale, was correlated with diagnostic psychiatric interviews (11). The CES-D is a measure of depressive symptoms composed of 20 self-report items, each with a Likert rating scale from 0 to 3. The maximum score is 60 (10). A score of> 16 has been shown to identify persons with depressive symptoms similar in severity to the levels observed among depressed patients (10,12,13). Open-ended questions provided respondents with the opportunity to relate the aspects of quarantine that were most difficult for them and allowed them to provide additional comments on their unique experiences.

Statistical Analysis

Means were calculated to summarize continuous variables. For categorical variables, group proportions were calculated. Student t tests were used to examine relationships between demographic variables and the psychological outcome variables, the scores on the IES-R and CES-D. A score of >20 on the IES-R was used to estimate the prevalence of PTSD symptoms (11). A score of >16 on the CES-D was used to estimate the prevalence of depressive symptoms (10,12,13).

Analysis of variance (ANOVA), chi-square, and the Cochran-Armitage test for trend were used to examine relations between the IES-R and CES-D scores and the following independent variables: healthcare worker status, home or work quarantine, acquaintance of or direct exposure to someone with a diagnosis of SARS, combined annual household income, and the frequency with which persons placed in quarantine wore their masks. Linear regression for the trends between income categories and both PTSD and depressive symptoms was analyzed. The relationships between the IES-R and CES-D and whether persons in quarantine wore their masks all of the time versus never were examined by the Duncan-Waller K-ratio t tests. A p value of < 0.05 was considered to be significant for all analyses.

Qualitative data were coded and analyzed to show emerging themes. The development and confirmation of the thematic coding structure is an iterative process involving two researchers in individual, recursive reading of the textual data and group meetings to discuss and test the emerging themes. Discrepancies were resolved by consulting specific instances in the data, discussing their relationship to established themes, and reaching consensus as a group (14).

Demographics and Description of Quarantined Persons

Figure

Figure. Number of persons in quarantine, Toronto, Canada, February 23–June 30, 2003. Figure courtesy of Toronto Public Health.

The survey was completed by 129 of more than 15,000 eligible persons who were placed in quarantine (Figure). All respondents completed the survey at the end of quarantine with a minimum time from the end of quarantine to the completion of the survey of 2 days. The median time from the end of quarantine to completion of the survey was 36.0 days (interquartile range, 10–66 days). Sixty-eight percent of respondents were healthcare workers, 64% were 26–45 years of age, 58% were married, 72% had a college level of education or higher, and 48% had a combined household income of >$75,000 (Canadian dollars [CAD]).

The 129 respondents described 143 periods of quarantine with 90% of respondents being placed into quarantine only once; 66% of respondents were on home quarantine, while 34% were on work quarantine. The median duration of quarantine was 10 days (interquartile range 8–10 days). Half of respondents knew someone who was hospitalized with SARS of whom 77% were colleagues; 10% knew someone who had died of SARS (Table 1).

Persons were notified of their need to go into quarantine from the following sources: their workplace (58%), the media (27%), their healthcare provider (7%), and public health officials (9%). Most (68%) understood that they were quarantined to prevent them from transmitting infection to others; 8.5% of respondents believed they were quarantined to protect themselves from infection; 15% did not believe they should have been placed into quarantine at all; and 8.5% provided more than one of these responses.

The source of notification for quarantine influenced understanding of the reason for quarantine. Those who were notified by the media or their workplace were more likely to understand the reason for quarantine than those who were notified by their healthcare provider or public health unit (p = 0.04). Healthcare workers were also more likely to understand the reason for quarantine compared with non–healthcare workers, 76.5% versus 52.5% (p = 0.007). Combined household income and level of education did not influence understanding of the reason for quarantine.

Information on Infection Control Measures

Persons received their information regarding infection control measures to be adhered to during their quarantine from the following sources: the media (54%), public health authorities (52%), occupational health department (33%), healthcare providers (29%), word-of-mouth (23%), hospital Web sites (21%), and other Web sites (40%).

Those who did not think they had been well-informed were angry that information on infection control measures and quarantine was inconsistent and incomplete, frustrated that employers (healthcare institutions) and public health officials were difficult to contact, disappointed that they did not receive the support they expected, and anxious about the lack of information on the modes of transmission and prognosis of SARS (Appendix).

During the outbreaks, nearly 30% of respondents thought that they had received inadequate information about SARS. With respect to information regarding home infection control measures, 20% were not told with whom they could have contact; 29% did not receive specific instructions on when to change their masks; and 40%–50% did not receive instructions on the use and disinfection of personal items, including toothbrushes and cutlery; 77% were not given instructions regarding use and disinfection of the telephone. Healthcare worker status did not influence whether respondents thought they had received adequate information regarding any of the listed home infection control measures, except regarding the frequency of mask changing: healthcare workers more frequently reported that they had received adequate information, 78.8% versus 60.5% (p = 0.03).

Adherence to Infection Control Measures

Eighty-five percent of quarantined persons wore a mask in the presence of household members; 58% remained inside their residence for the duration of their quarantine. Thirty-three percent of those quarantined did not monitor their temperatures as recommended: 26% self-monitored their temperatures less frequently than recommended, and 7% did not measure their temperatures at all. No differences between healthcare workers and nonhealthcare workers were found with respect to adherence to recommended infection control measures.

Psychological Impact of Quarantine

The mean IES-R score was 15.2±17.8, and the mean CES-D was 13.0±11.6. The IES-R score was >20 for 28.9%; the CES-D score was >16 in 31.2% of quarantined persons (Table 2). The mean IES-R scores were not different for persons on home or work quarantine, 14.1±18.8 versus 17.6±16.6 (p = 0.33); the mean CES-D scores were also not different between the groups, 12.0±12.0 versus 15.2±10.7 (p = 0.16).

The presence of PTSD symptoms was correlated with the presence of depressive symptoms (p < 0.0001, r = 0.78). Marital status did not offset the presence of PTSD symptoms, mean IES-R score of 14.5±16.6 for those who were unmarried versus 13.8±14.6 for those who were married (p = 0.82). Similarly, marital status did not influence the presence of depressive symptoms, with a mean CES-D score of 12.9±10.7 for those who were unmarried versus 12.5±11.4 for those who were married (p = 0.85)

A combined annual household income of CAD <$40,000 versus CAD $40,000 to CAD $75,000 versus CAD >$75,000 was associated with increased PTSD symptoms (mean IES-R score of 24.2±20.6 versus 20.0±24.4 versus 11.8±11.6, respectively) (p = 0.03 for the three-way comparison). Linear regression testing for trend over income categories was also significant (p = 0.01). A combined annual household income of CAD <$40,000 versus CAD $40,000 to CAD $75,000 versus CAD >$75,000 was also associated with increased depressive symptoms (mean CES-D score of 18.3±15.4 versus 15.5±13.2 versus 10.9±9.2, respectively) (p = 0.05 for the three-way comparison) (Table 2). Results of linear regression testing for trend over income categories were also significant (p = 0.01).

Neither age, level of education, healthcare worker status, living with other adult household members, nor having children was correlated with PTSD and depressive symptoms. The duration of quarantine was significantly related to increased PTSD symptoms, mean IES-R score of 23.7±27.2 for those in quarantine >10 days compared with 11.7±10.7 for those in quarantine <10 days (p < 0.05). Persons who were in quarantine for a longer duration showed a trend toward higher CES-D scores; however, this difference did not reach statistical significance (mean CES-D of 17.0±14.2 for those in quarantine >10 days versus 11.2±10.1 for those in quarantine <10 days [p = 0.07]). Acquaintance with or exposure to someone who was hospitalized with SARS was associated with a higher mean IES-R score, 18.6±20.2 versus 11.8±14.3 (p = 0.03) and a higher mean CES-D score, 15.5±12.1 versus 10.2±10.5 (p = 0.01). Overall, acquaintance with or exposure to someone who died of SARS was not correlated with PTSD or depressive symptoms (data not shown).

Persons were categorized as having worn their masks all of the time, including times when it was not recommended, having worn their masks according to recommendations, or not having worn their masks at all. Those who wore their masks all of the time had higher mean IES-R scores (29.7±18.6 versus 14.1±17.9 versus 12.3±15.1, p = 0.003 for the three-way comparison) and higher mean CES-D scores (25.6±12.7 versus 12.2±11.1 versus 11.5±11.6, p = 0.002 for the three-way comparison). Those who wore their masks all of the time also had higher mean IES-R scores (p = 0.03) and higher mean CES-D scores (p = 0.002) than those who never wore their masks.

All respondents described a sense of isolation. The mandated lack of social and, especially, the lack of any physical contact with family members were identified as particularly difficult. Confinement within the home or between work and home, not being able to see friends, not being able to shop for basic necessities of everyday life, and not being able to purchase thermometers and prescribed medications enhanced their feeling of distance from the outside world. Infection control measures imposed not only the physical discomfort of having to wear a mask but also significantly contributed to the sense of isolation. In some, self-monitoring of temperature provoked considerable anxiety: "taking temperatures was mentally difficult" (respondent #27) and "taking my temperature made my heart feel like it was going to pound out of my chest each time" (respondent #62).

While most quarantined persons (60%) did not believe that they would contract SARS, 59% were worried that they would infect their family members. In contrast, only 28% were concerned that a quarantined family member would infect someone else in the home. Following quarantine, 51% of respondents had experiences that made them feel that people were reacting differently to them: avoiding them, 29%; not calling them, 7%; not inviting them to events, 8%; and not inviting their families to events, 7%.

Comments from survey respondents

Unmet informational needs:

1. Public health /employers:

a. Difficulty in access: "Called Public Health for 2 days. Got through 3 times; waited on hold for hours, then got hung up on." (respondent # 131)

b. Failed expectations: "I was expecting someone from Public Health to check up on me but never got a call except on my last day of quarantine." (respondent #126); "Nobody told me anything. I was not contacted by health officials at all." (respondent# 99); "My employer should have been more forthcoming." (respondent #7); "I was not called by the hospital I worked at. I saw the quarantine on the news and spent a whole day trying to get through to my unit." (respondent #40)

c. Lack of support: "I was looking for more support from the health care professionals. They left me in the dark to deal with this." (respondent #22)

2. Nature of information:

a. Details re: infection control: "I have since learned that there are a lot of precautions that no one ever told me about." (respondent #81)

b. Inconsistencies: "Information was not always the same. Many inconsistencies." (respondent #66)

c. Timing: "Information was given too late, as I started 1 week after exposure. Unacceptable!" (respondent #27)

d. Specific issues:

i. Children: "Nobody can tell me exactly where my children would be arranged to go in case I got SARS myself. I was very panicked at that time and my husband was admitted that time because of the SARS." (respondent # 78)

ii. Onset of symptoms: "What symptoms were considered serious and what to do when I experienced those symptoms." (respondent # 21); "I was mildly alarmed to realize that I didn't know what to do if I actually did develop symptoms of SARS." (respondent # 111)

iii. Prognosis of SARS: "Most of the really important info is largely unknown" (respondent #53); "Prognosis for SARS, how many have recovered, what health problems recovered patients still have." (respondent #8I)

iv. Mode of transmission: "If airborne what were the chances of contracting the disease… MD unable to answer." (respondent #90)