SEARCH

SEARCH BY CITATION

Keywords:

  • actigraphy;
  • electronic diaries;
  • infant sleep;
  • paper diaries

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Reliable, valid and cost-effective methods for the assessment of infant sleep and sleep problems are of major importance. In this study, the first aim was to assess the agreement of an electronic diary as well as a paper diary with actigraphy for measuring infant sleep patterns in a community sample. The second aim was to assess the feasibility and acceptance of, and compliance with, the electronic diary and the paper diary. Ninety parents reported infant sleep behavior in a paper diary in their home environments for a total of 6 days, 95 in an electronic diary, within two consecutive weeks while actigraphic data were obtained simultaneously. We found moderate to good agreement between electronic diaries and actigraphy (r = 0.41–0.65, P < 0.01), and paper diaries and actigraphy (r = 0.47–0.70, P < 0.01). In addition, this study also found good agreement between both diaries and also between both diaries and actigraphy for sleep percentage over 24 h (electronic diaries and actigraphy: 54.1 ± 0.7%, 52.5 ± 0.7%, P < 0.05; paper diaries and actigraphy: 55.1 ± 0.5%, 52.2 ± 0.6%, P < 0.01) and for daytime (electronic diaries and actigraphy: 27.3 ± 0.9%, 23.5 ± 1.2%, P < 0.01; paper diaries and actigraphy: 27.3 ± 0.8%, 23.2 ± 1.0%, P < 0.01), with the exception that less daytime sleep was recorded on actigraphy than on either diary. In conclusion, the electronic diary and the paper diary are valid and well-accepted methods for the assessment of infant sleep. Parents preferred the electronic diary but, conversely, they were less compliant in completing it.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Development of infant sleep is one of the most important physiological processes that occurs over the first year of life. Sleep has been described as a ‘window into the developing brain’ (Kohyama, 1998), and thus the assessment of the development of normal sleep has important implications for understanding later behavioral problems.

Assessments of infant sleep need to be able to be carried out in a cost-effective manner with as little disruption to infant/parent daily routines as possible.

To date, a variety of research methods have been used in clinical and community-based samples. Laboratory polysomnography – the most reliable and comprehensive method for assessing infant sleep (Ednick et al., 2009) – is limited by the short period of assessment and the artificial sleep environment, which may be disruptive to sleep. Actigraphy is an alternative method for the assessment of infant sleep. It allows continuous, non-invasive recordings in the natural environment over prolonged periods without time-consuming home installation procedures or restriction to one specific sleep setting (e.g. the child’s crib). The validity of actigraphy has been demonstrated in studies with polysomnography (Ancoli-Israel et al., 2003; Sadeh et al., 1995a).

Parental diary reports are an alternative to the previous methods and are commonly used for assessing infant sleep (Bolger et al., 2003; Thiele et al., 2002). Barr et al. (1988) developed two versions of a parental diary of infant and care-giver behaviors, including infant sleep; one is a paper diary and one is an electronic version of the same diary (electronic diary, Hunziker and Barr, 1986; Lam et al., 2010).

The aim of the current study was to include an electronic diary in the comparison of sleep measurement methods. We compared sleep times collected with both electronic diaries and paper diaries in each infant, and compared these with concurrent actigraphic measurements of sleep. An additional aim was to assess the feasibility and acceptance of, and compliance with, the electronic diaries and paper diaries in a relatively large-scale prospective study. We expected good agreement among the different methods of assessment, and good acceptance of, and compliance, with both diaries.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

The study was conducted as a sub-project of the Swiss Etiological Study of Adjustment and Mental Health (SESAM) study at the Department of Clinical Child and Adolescent Psychology, University of Basel, Switzerland. The Ethical Committee of Basel approved all procedures. Mothers were included when German was their first language, and when their child was between the ages of 1 and 12 months.

Subjects

Initially 167 parents and infants were recruited. The final sample for the comparison of electronic diary with actigraph consisted of 90 healthy infants (45M/45F), and for the comparison of paper diary with actigraph of 95 healthy infants and their parents (50M/45F; Fig. 1). For the 95 participants, the mothers’ mean age was 32.0 ± 3.8 years with an average of 16.8 ± 3.1 years of education. The majority of parents were Swiss (55.8% of mothers; 53.7% of fathers). Eighty-seven mothers were living in a stable relationship. The fathers’ mean age was 34.4 ± 5.4 years. The infants’ age was between 1 and 9 months with a mean age of 5.0 ± 2.0 months. Infants were born at 39 ± 0.2 weeks with mean birth weights of 3311 ± 64 g, mean birth lengths of 50 ± 0.3 cm, and an Apgar score of 9.

image

Figure 1.  Flow chart of the study.

Download figure to PowerPoint

Study protocol

Parents were contacted by telephone, and the 2-week assessment was scheduled. Parents received a package by mail consisting of electronic diaries, paper diaries, actigraphs (MicroMini Motionlogger Actigraph; Ambulatory Monitoring, Ardsley, NY, USA) and a recording sheet of infant sleep behavior that was used to facilitate the actigraphy analysis, personal instructions and questionnaires assessing maternal psychopathology. Parents were called several times, instructed on how to place actigraphs on the child’s ankle, and how to fill out diaries. Infants wore actigraphs for three consecutive and continuous days in the first and second week of assessment simultaneous to parental recording in the diaries.

Parents completed the electronic diary (Palm-type personal digital assistant, PDA; Palm Tungsten E2 with customized software written by Hiloma Software Development, Montreal, QC, Canada; see Lam et al., 2010) and the paper diary developed by Barr et al. (1988). For the electronic diary, parents recorded infant behavior and their own care-giving behavior for periods of 24 h with predetermined symbols. Parents were asked to indicate on the touch screen the relevant infant behavior from the beginning to the end of the duration of the specific behavior. Parents filled in diaries every few hours when it was possible and convenient for them. The data were saved and processed automatically, and then downloaded to a computer. Electronic diaries were not time-stamped.

The paper diary was developed by Barr et al. (1988), and records the same infant and care-giver information as that obtained with the electronic diary. Data were analysed for each infant if parents had completed data entry for at least 50% of each of the 3 days of study. In case of data loss of more than 50% of any 1 day, that entire day was excluded from data analyses.

At the end of each recording period, the diary’s acceptability was assessed with acceptance questionnaires on a five-point categorical scale regarding how troublesome and time-consuming the diary was in the family’s normal routine (rating responses: 1 = not at all; 2 = a little; 3 = somewhat; 4 = very; 5 = extremely). After the end of both recording periods, participants were asked to assess their preference for the electronic diary or the paper diary on the same five-point categorical scale, and rated the extent of their preference for the method they preferred.

Data analysis

A period of 24 h was defined as 00:00–24:00 hours; daytime was defined as 08:00–20:00 hours, and night-time was defined as 20:00–8:00 hours, based on the study of So et al. (2007). Paper diaries and electronic diaries were analysed manually, and data were aggregated in sleep/wake minutes. Actigraph data were analysed with the validated Sadeh actigraph infant scoring algorithm (ASA, Actigraphs Sleep Analysis; Sadeh et al., 1989) and coded into sleep/wake minutes. Electronic diary, paper diary and actigraphy were matched for the total study duration and for the comparison of electronic diary and actigraphy as well as paper diary and actigraphy. The total percentage of sleep over each 24-h period, during the day and at night was calculated for all methods of recording. First, correlations between electronic diary and actigraphy, paper diary and actigraphy, and electronic diary and paper diary were derived with Pearson’s correlation coefficient r. Second, agreements between electronic diary and actigraphy, and paper diary and actigraphy were evaluated according to the method of Bland and Altman (1986). Third, the differences of the three methods of assessment regarding total percentage of sleep over each 24-h period were compared with paired Student’s t-tests. Differences between the groups for gender, number of siblings, age of the child, maternal psychopathology and income were compared with non-paired Student’s t-test.

Differences were considered statistically significant if P < 0.05. The duration of missing data of all completed electronic diaries and paper diaries was compared using paired Student’s t-tests. Acceptability ratings were compared using paired Student’s t-tests. Exact binomial tests were used to analyse preferences of electronic diaries and paper diaries.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Ninety data sets were included for the comparison of electronic diary and actigraphy, and 95 data sets for the comparison of paper diary and actigraphy. Ninety different, only partially overlapping, data sets were included for ratings on acceptance with and on preference for electronic diaries and paper diaries. Participants assessed their preference for the electronic diary or the paper diary on a five-point categorical scale, and rated the extent of their preference for the method they preferred.

Validation of sleep duration by electronic diary and paper diary against actigraphy

Table 1 presents correlations of sleep percentage between electronic diary and actigraphy, paper diary and actigraphy, and electronic diary and paper diary. Over 24 h, during the day and at night, we found medium to large correlations between electronic diary and actigraphy (r = 0.41–0.65, P < 0.01), between paper diary and actigraphy (r = 0.47–0.70, P < 0.01), and between electronic diary and paper diary (r = 0.57–0.73, P < 0.01).

Table 1.   Correlations between e-diary and actigraphy, paper diary and actigraphy, electronic diary and paper diary of sleep percentage over 24 h (00:00–00:00 hours), during the day (08:00–20:00 hours) and at night (20:00–08:00 hours)
 Electronic diary and actigraphyPaper diary and actigraphyElectronic diary and paper diary
  1. **P < 0.01.

Sleep percentage over 24 hr = 0.41**r = 0.57**r = 0.57**
Sleep percentage during the dayr = 0.65**r = 0.47**r = 0.73**
Sleep percentage at nightr = 0.64**r = 0.70**= 0.72**

Fig. 2 presents the agreement between actigraphy and electronic diary, as well as actigraphy and paper diary. Over 24 h, paper diary but not electronic diary overestimated time asleep compared with actigraphy according to the method of Bland and Altman (1986).

image

Figure 2. Bland and Altman (1986) plots depict: (a) the difference in sleep percentage of actigraphy and electronic diary (y-axis) versus mean sleep percentage of actigraphy (x-axis) over 24 h; (b) the difference in sleep percentage of actigraphy and paper diary versus mean sleep percentage of actigraphy over 24 h. Data are presented as mean ± 2 SD.

Download figure to PowerPoint

Fig. 3 presents sleep percentage recorded by electronic diary, paper diary and actigraphy. Over 24 h, electronic diary indicated significantly more time asleep compared with actigraphy (54.1 ± 0.7% versus 52.5 ± 0.7%, P < 0.05), and paper diary compared with actigraphy (55.1 ± 0.5% versus 52.2 ± 0.6%, P < 0.01). During the day, electronic diary indicated significantly more time asleep compared with actigraphy (27.3 ± 0.9% versus 23.5 ± 1.2%, P < 0.01), and paper diary compared with actigraphy (27.3 ± 0.8% versus 23.2 ± 1.0%, P < 0.01). At night, no significant differences were found between electronic diary and actigraphy (82.7 ± 0.8% versus 82.5 ± 0.9%, P = 0.78), or between paper diary and actigraphy (83.3 ± 0.8% versus 83.4 ± 0.8%, P = 0.78).

image

Figure 3.  Overall percentage of time spent asleep over 24 h (00:00–00:00 hours), during the day (08:00–20:00 hours) and at night (20:00–08:00 hours), as recorded by e-diary (dark gray bars), p-diary (black bars), actigraphy (actigraphy assessment simultaneously to e-diary report, light gray bars), and actigraphy (actigraphy assessment simultaneously to p-diary report, white bars).

Download figure to PowerPoint

Validation of sleep duration by electronic diary and paper diary against actigraphy at different ages

The total sample was divided into two age groups (1 = 0–6.0 months; 2 = 6.1–12 months).

Over 24 h, electronic diary indicated significantly more time asleep compared with actigraphy for both age groups, and paper diary compared with actigraphy for both age groups (P < 0.05; Fig. 4).

image

Figure 4.  Mean percentage total time spent asleep assessed with e-diary, p-diary and actigraphy during 24 h for two age groups (infants until the age of 6 months, infants above 6 months), recorded by e-diary (dark gray bars), p-diary (black bars), actigraphy (actigraphy assessment simultaneously to e-diary report, light gray bars), and actigraphy (actigraphy assessment simultaneously to p-diary report, white bars).

Download figure to PowerPoint

Differences of sleep duration assessed by diaries against actigraphy for parental/child characteristics

No differences were found for relationship status, number of siblings, maternal psychopathology and income.

At night, paper diary indicated significantly less time asleep for females compared with males when compared with actigraphy (81.6 ± 1%, 85.1 ± 1.2%, P < 0.05).

Feasibility of electronic diary and paper diary

Feasibility referred to technical and organizational difficulties leading to data loss. Eleven (12%) complete electronic diaries had to be excluded because of electronic diary malfunctions, and four (4%) complete electronic diaries were lost because of power problems (technical problems).

Acceptance with and preference for electronic diary and paper diary

Table 2 gives an overview of diary acceptance and diary preference ratings based on post-diary questionnaires. Parents found the electronic diary less troublesome and less time-consuming, and also preferred the electronic diary compared with the paper diary.

Table 2.   Parental ratings on acceptance with and on preference for electronic diaries and paper diaries based on the following rating responses: 1 = not at all; 2 = a little; 3 = somewhat; 4 = very; 5 = extremely
 Electronic diary Mean SDPaper diary Mean SD
  1. *P < 0.05, **P < 0.01.

Filling out the diary was
 Troublesome1.5* (1.0)n = 901.9* (1.1)n = 95
 Time-consuming2.0** (1.2)n = 902.6** (1.4)n = 95
Preference for3.3** (1.8)n = 451.7** (1.4)n = 45

Compliance with electronic diary and paper diary regarding completion of diary data

Compliance was defined in relation to parental completion of diary entry and not in terms of the timing of data entry. That is, the extent to which participants followed the instructions and completed each of the 3 days of diary recording and at least 50% of diary recording on any 1 day in each diary. Data were included when there was <50% of each day completed. In case of data loss of more than 50% of any 1 day, that entire day was excluded from data analyses.

Nine (10%) electronic diaries and three (3%) paper diaries had to be excluded because of parental non-compliance (parents did not complete at least 50% of diary recordings for all three electronic diary or paper diary days). Twenty-four (25%) of the 95 paper diaries included in the final sample had missing data (<50%) for 1 day, six (6%) for 2 days and four (4%) for 3 days. Thirty (33%) of the 90 electronic diaries included in the final sample had missing data (<50%) for 1 day, 25 (28%) for 2 days and 10 (11%) for 3 days.

Participants using electronic diaries showed significantly more missing data (in minutes) for all diaries and all 3 days included with a mean of 40.7 ± 77.7 min for electronic diaries, and with a mean of 30.0 ± 67.7 min for paper diaries, P < 0.01.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Validation of sleep duration by electronic diary and paper diary against actigraphy

This was the first reported validation study comparing both electronic diaries and paper diaries against actigraphy for sleep behavior in infants. There were two main findings. First, we found medium-to-large significant correlations between electronic diary and actigraphy, and paper diary and actigraphy over 24 h, during the day and at night. Over 24 h and during the day, electronic diary and paper diary indicated significantly more time asleep compared with actigraphy. Over 24 h, paper diary overestimated time asleep according to the method of Bland and Altman (1986).

Previous validation studies reported good correlations between parental paper diary reports and actigraphy (Sadeh, 1994, 1996; So et al., 2007). Sadeh (1994, 1996) found strong correlations between paper diaries and actigraphy for sleep duration at night in infants with night-waking problems. To our knowledge, only one community study compared the parental paper diary report with actigraphy in infants (So et al., 2007). So et al. followed 20 infants longitudinally, and assessed sleep/wake patterns with paper diary and actigraphy in the infants’ home environments for 3 days each month for the first 12 months of life. The results showed overall agreements between paper diary and actigraphy for 24-h recordings, 12-h recordings during the day, and at night. Significant differences were found also at night with a higher percentage of sleep time for paper diary compared with actigraphy (80.7 ± 1.0% versus 73.3 ± 0.9%, P < 0.01). This result is of particular interest because at night we found, instead, a similar percentage of sleep time for actigraphy compared with electronic diary.

A possible explanation for this result could be related to the study designs. So et al. (2007) followed 20 healthy infants during their first year of life and compared paper diaries with actigraphy longitudinally. They assessed infant sleep over three consecutive days each month. Sadeh (1994, 1996) included infants with night-waking problems in his studies.

Validation of sleep duration by electronic diary and paper diary against actigraphy at different ages

We found that electronic diary and paper diary indicated significantly more time asleep compared with actigraphy for both age groups. These results are in line with the results of So et al. (2007) who also reported that paper diary indicated significantly more time asleep at the age of 2, 5, 9 and 11 months (24 h) compared with actigraphy. In contrast, So et al. (2007) reported that paper diary indicated significantly less time asleep compared with actigraphy at 1 month. Actigraphy has been shown to be less reliable in younger infants (Gnidovec et al., 2002; Sadeh et al., 1995b; So et al., 2005), and the specific kind of actigraph used also might have an influence on the results.

Differences of sleep duration assessed by diaries against actigraphy for parental/child characteristics

At night, paper diaries indicated significantly more time asleep compared with actigraphy scored for females. This result is in line with the results of So et al. (2007) who reported that significantly more sleep minutes were scored during the day for females at 1 month for paper diary and actigraphy. At 2 months actigraphy again indicated significantly more time asleep for females.

Feasibility of electronic diary and paper diary

We lost 16% (n = 15) of electronic diary data due to technical difficulties. Previous studies also reported technical difficulties with electronic diaries such as battery problems and electronic diary malfunction (Lauritsen et al., 2004; Nyholm et al., 2004; Palermo et al., 2004; Tiplady et al., 1997). Our results are in line with these studies that showed data loss ranging between 3.5 and 14% (Nyholm et al., 2004; Palermo et al., 2004; Tiplady et al., 1997).

The study took place by phone and without any face-to-face contact. All materials including electronic diaries and actigraphs were sent by mail to the parents, completed by the parents themselves, and sent back after the end of the study – in general after 3–4 weeks. In particular, the lack of face-to-face contact, and thus the lack of the ability to develop a mutual relationship between the research team and participants, probably influenced participants’ compliance.

Parents and infants in this study lived throughout Switzerland and Germany, so that the geographical distance limited fast and immediate support possibilities including repairing, fixing or changing electronic diaries and actigraphs. On the one hand, technical failures and difficulties such as electronic diary malfunctions, power problems or battery problems appeared to be large in our study; on the other hand, additional limitations that hindered the compensation of these difficulties explained this percentage of data loss.

Electronic diaries have advantages for data handling and data management (Nyholm et al., 2004; Tiplady et al., 1997). Electronic diaries and portable technologies comply with today’s Western world standards. On the other hand, the monetary and time expenses for electronic diaries seem to be higher than for paper diaries. The monetary expenses not only include the costs for the Palms, and the costs for the acquisition and development of the software (Dale and Hagen, 2007), but the price of batteries, the costs of insuring, repairing or replacing the Palms (Christensen et al., 2003).

The research team conducting the study has to be well trained and has to obtain technical expertise in this specific area. Our study, as well as the majority of studies comparing electronic diaries with paper diaries, reported technical problems that often led to data loss. This major drawback of electronic diaries has to be taken into account carefully because it can correspond with potentially high monetary and time expenses. Also, the research team has to be able to respond and cope with any technical problems and support questions in a very flexible, continuous and ongoing manner over the whole period of assessment.

Acceptance with and preference for electronic diary and paper diary

Parents found the electronic diary less troublesome and less time-consuming compared with the paper diary. Parents preferred the electronic diary. This result is in line with previous studies that found that participants preferred the electronic diary compared with the paper diary (e.g. Heiberg et al., 2007; Tiplady et al., 1997). Preference for, and higher satisfaction with, the electronic diary is probably related to advantages regarding portability and data entry. The majority of people own a mobile phone and are experienced with the use of portable technology, which again might be related to participants’ preference for electronic diaries.

Compliance with electronic diary and paper diary regarding completion of diary data

We found lower compliance for electronic diary data completion in our study. This result is different from previous studies that generally supported the notion that the electronic diary improves protocol compliance (Nyholm et al., 2004; Palermo et al., 2004; Rabin et al., 1993, 1996; Stone et al., 2003). Only one study reported better protocol compliance for paper diary (Lauritsen et al., 2004). Protocol compliance in the studies was defined in relation to the timing of data entry of the participants within a designated time frame (e.g. time frame of 15 min). All studies relied on time-stamps for the assessment of electronic diary compliance.

In our study, we defined compliance as the extent to which participants comply with the instructions to fill in the different codes for child and parental behavior for all times, with no regard to the timing aspect. We speculate that lower maternal compliance for electronic diaries was the result of electronic diary malfunction and technical problems leading to maternal fatigue and frustration regarding electronic data entry. Additionally, it is likely that the lack of regular face-to-face contact affected participants’ compliance negatively.

Limitations

The electronic diaries used in our study were not time-stamped. According to the studies of the research group of Barr et al. (1988); (Lam et al., 2010), we instructed the parents to fill in the diaries every few hours, when it was convenient for them, without disruptions of their daily activities. Thus, we do not have information on how often and when parents filled out the data, independent of paper and electronic diary. According to the results of Lam et al. (2010), however, we suppose that daily electronic diary entries tend to decline across weeks and settle at two–three entries a day when parents are instructed to fill in the diaries when convenient for them.

Diaries were completed for 3 days at a time. This time period corresponds with the typical time ranges used in research and in clinical settings for the assessment of infant sleep. Nevertheless, our findings, especially with regard to feasibility and compliance aspects, may have been different if diary completion had taken place for a shorter or longer time period.

Another limitation refers to implementing the study based on phone contact only. In contrast to previous infant sleep studies, we did not have face-to-face contact with the families. Thus, parents carried out the study themselves, and we relied on parental compliance for putting on actigraphs and for filling out the diaries adequately. On the other hand the present study demonstrates that even without face-to-face contact the implementation of diary studies is feasible and can be used in large-scale studies.

We only concentrated on the assessment of estimates of sleep time without validation of additional aspects of sleep variables (e.g. night-waking). Additional aspects were not included because the validation and feasibility of both diaries was the main focus of our study. Also, diaries and actigraphs only allowed measuring and comparing specific aspects of sleep.

We recruited 167 participants, and 116 participants were eligible to take part in the study. The samples consisted of 95, respectively 90, participants. Mainly technical problems regarding electronic diaries and actigraphy led to an additional decrease in the number of participants and to data loss.

Our study consisted of mainly highly educated families, and thus the results cannot necessarily be generalized to other sectors of the population.

Conclusion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Our study found moderate to strong agreements between electronic diary and actigraphy as well as paper diary and actigraphy over 24 h, during the day and at night. Electronic diary and paper diary indicated significantly more time asleep compared with actigraphy. Sixteen percent of electronic diary data was lost due to technical problems. In conclusion, electronic diary and paper diary are valid and well-accepted methods for the assessment of infant sleep. Parents preferred the electronic diary but, conversely, they were less compliant in completing it.

There is an ongoing debate about the superiority of electronic diaries versus paper diaries, especially regarding the choice of one method of assessment for studies. The results of the present study indicate that both diaries are feasible and valid methods for the assessment of infant sleep in large-scale studies. Thus, according to our opinion, additional aspects such as overall expenses, purpose and requirements of the study have to be taken into account (Lam et al., 2010). We suggest that a comprehensive cost analysis that includes monetary and time expenses is a necessity for the choice of which diary will be used for future studies. Studies that include electronic diaries should be able to guarantee well-prepared technical set-up, quality assurance of the electronic diary software, as well as technical expertise and competence utilizing additional compliance-enhancing features, additional visual instructions such as DVDs and face-to-face contact with the families.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

This work is part of the National Centre of Competence in Research (NCCR) Swiss Etiological Study of Adjustment and Mental Health (sesam). The Swiss National Science Foundation (SNF; project no. 51A240-104890), the University of Basel, the F. Hoffmann-La Roche Corp., and the Freiwillige Akademische Gesellschaft provided core support for the NCCR sesam. This publication is the work of the authors Silvana Müller, Mirja H. Hemmi, Frank H. Wilhelm, Ronald G. Barr and Silvia Schneider, who serve as guarantors for the paper.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References