988368

1. Introduction
One of the major occupations of children is play [1]. On the other hand, it is an everyday occupation performed within a setting. It is affected by internal and external personal or environmental items [2]. In addition, these elements are affected by sensory processing in children [3]. Sensory processing is a phrase used due to Ayres’ research in 1972-1979. This term refers to the processing and organizing of sensations during activities [3]. Evidence shows that children’s participation and choices in various activities of daily living are affected by sensory processing patterns [4, 5]. Therefore, having information about sensory processing science can help healthcare providers to pay attention to the children’s participation in different activities and plan an intervention for children with different kinds of impairments [6]. Accordingly, it is necessary to contemplate children’s sensory priorities when presenting games, leisure, and academic activities. To achieve this goal, we need to know how children’s participation in leisure and play can be increased or decreased by their sensory preferences.
To clarify the systematic search for articles, participation in leisure, play, and sensory processing was defined. In this study, the interpretation of the play was as follows: 1) Having an inner motivation to do the activity, 2) being free from external laws, 3) going beyond reality and also reflecting on being realistic, 3) focusing on the process of doing the activity instead of the result of the activity, 4) to be spontaneous and safe, and 5) to need the active participation of the player [7, 8]. Based on the the occupational therapy practice framework (OTPF), leisure time is a non-engaging and self-stimulate activity that is performed at an optional time when it is not devoted to mandatory occupations like doing self-care tasks or jobs [9]. Approximately 30% of a day devotes to voluntary activities in normal children [10]. Accordingly, the main part of their daily lives in children is devoted to playing and leisure activities.
Dunn developed a sensory processing model that proposes an association between an individual’s neural system and self-regulatory techniques [4]. In Dunn’s model, the spectrum of neural thresholds is crucial. For example, strong input is not demanded by people with lower thresholds for sensory stimuli. However, they often do not notice the stimulus if they have a high point. Another major concept is self-regulation, in which individuals can respond passively or actively to sensory stimuli. In the intersection of these two spectrums, Dunn presents four patterns of sensory processing: Sensory seeking, sensory avoiding, sensory sensitivity, and low registration.
According to the definition of the World Health Organization (WHO), participation is an essential part of human development, through which proficiencies, qualifications, and the meaning of life are acquired [11]. Therefore, to optimize children’s participation in society, it is necessary to find out how their sensory processing ability affects their participation in daily life. This study aims to raise the therapists’ awareness of the mentioned cases. Accordingly, a systematic review of articles was carried out to investigate the association between sensory processing patterns and children’s participation in leisure and play activities. The main research question is, “What is the association between sensory processing and participation in leisure activities and play in children aged 3 to 14 years?” Hence, this systematic review seeks to explore the following issues: The research undertaken on this subject so far, the quality of existing evidence, and the pooled measure of the correlation between the patterns of sensory processing and participation in leisure activities.

2. Materials and Methods
The PRISMA (the preferred reporting items for systematic reviews and meta-analyses) guidelines were followed in conducting this systematic review [12]. First, to review the literature on this subject, we searched Persian and equivalent English keywords such as “participation”, “leisure”, “entertainment”, “play”, “sensory processing”, “sensory integration”, “sensory”, “motor sensory”, “occupational therapy” and “children”, in online databases like PubMed, Google Scholar, Springer, ProQuest, Scopus, Cochrane Library, Medline OT seeker, SID, and Healio.
To define texts that can be included in this study, play, and leisure activities were defined broadly. For example, play denotes the sensual and physical discovery of playthings and the settings, manipulation of playthings with or without friends, and spontaneous engagement in children. The play consists of capabilities like imagination, management of playthings, social communication, or fine motor dexterities.
Other inclusion criteria of the papers are as follows:
1. Collection and evaluation of data about play or leisure and sensory processing, 
2. Publication in a reliable journal, 
3. Description of data collection or evaluation method in the text. 
4. The sample consisted of children aged 3-14 years, 
5. The research published during 2010-2021, and 
6. The availability of the full text (English or Persian). 
Studies that did not include a full text did not examine children aged 3-14 years, did not examine both sensory processing and leisure or play participation, and were written in languages other than English or Persian were excluded. In addition, studies that examined involvement in activities other than leisure and play were excluded.
The study selection process was carried out in 2 phases, all titles were evaluated, and those not compatible with the study’s objectives were excluded. All articles collected through the primary search were listed, and their titles, abstracts, and references were included in EndNote software, version 20. Duplicates were removed automatically and then manually. Two reviewers then checked the titles and summaries of the articles on sensory processing and leisure or play participation. The final list of the selected articles was drawn up and, in the event of controversy on the inclusion or exclusion of an article, the agreement was achieved after the discussion between the reviewers. If no consensus was reached, the study went into a second phase in which the full text of the article was studied and examined independently by the two reviewers. Next, they discussed later disputes, and if there was no agreement, the opinion of a third reviewer was sought. In this way, the final list was created Table 1.

The process of identifying articles is shown in Figure 1.

Statistical analysis
The meta-analysis of the data, such as the association between sensory processing and participation in leisure activities and play, was conducted using the Mantel-Haenszel method in comprehensive Meta-analysis software (Version. 2.2; Biostat, Englewood, NJ, USA). The degree of heterogeneity between the results of the included studies was evaluated with the I2 index, which indicates the proportion of variability between studies, and, if statistically significant, a random effect was used as the basis for the meta-analysis. Funnel plots were used to present the results of the complete meta-analysis to investigate a possible publication bias in the studies.

Critical evaluation and evidence grades
The checklist of Joanna Briggs Institute (JBI) was employed to measure the quality of studies. The quality evaluation results are outlined in Table 2.

Also, the JBI checklist for case-control, cross-sectional, and systematic review studies is attached in Appendix 1. The included studies were then assigned to different grades of evidence. For this purpose, the grades of evidence were introduced by the National Health and Medical Research Council of Australia [12].

3. Results
According to the JBI checklist, a total score of more than 80% is defined as high quality, a score between 60% and 80% as medium quality, and a score of less than 60% as low quality [13]. None of the articles were removed because they were of accep quality. Of 13 examined articles, 11 had high quality, and 2 had medium quality.
The bulk of the research was undertaken in the United States (seven studies) and the rest in Australia (two studies), Malaysia (one study), and Israel (three studies).

Study groups
The studies were performed on children with autism spectrum disorder (ASD) (five articles), sensory processing disorder (two articles), normally developed children with potential sensory processing disorders (one article), children with attention-deficit hyperactivity disorder (ADHD) (one article), atypical dermatitis (one article), and normally developed children (two articles).

Articles on children with autism spectrum disorder
Preliminary research exhibited a significant relationship between the sensory system’s functions and played in children with ASD [14] In three studies, this association was investigated based on the type of disorders in sensory processing patterns. In Hochhauser and Engel-Yeger ‘s study [15], a significant correlation was reported between enhanced sensory processing ability and greater intensity of participation in children. Accordingly, the observed significant correlations between the desire to participate in leisure activities and sensory processing disorders were as follows:
1. High tactile sensitivity was associated with the increased intensity of participation in physical activity,
2. High taste/smell sensitivity was associated with low participation intensity,
3. Taste/smell sensitivity was associated with the delight of performing recreational activities with others, 
4. High motor sensitivity was related to several activities, particularly recreational and informal activities at home
5. A high sensory search was related to the performance of more activities, especially self-improvement activities at home, and Lower energy levels in children were related to the tendency to do activities with others [15].
Reynolds et al. also reported that increased sensory sensitivity and sensory avoidance significantly correlated with lower activity and social competence in children [16]. Then again, Little et al. stated that sensory hyper-responsiveness significantly affects neighborhood-social activities and the routine work (work done at home) in children and encourages outdoor activities. In addition, children with a greater tendency for sensory interests, repetitive movements, and sensory seeking preferred joint activities with other children and parents to alfresco activities [17].
In the study of Loh et al., sensory processing was found to be significantly correlated with participation in children’s occupations. Also, more problems in sensory processing were related to increased obstacles to participation. The auditory filtering was the only component of sensory processing significantly associated with participation problems (r=0.36, P=0.01), participation intensity (r=0.22, P<0.05), and participation enjoyment (r=0.27, P<0.01) [18].

Articles on children with sensory processing disorder
As asserted by Cosbey et al., children with sensory processing disorder (SPD) differ from their typical peers in terms of the intensity of participation and the people with whom they are engaged. Moreover, typical children have more diverse social networks in their social activities than children with SPD [19]. In a 2012 study, the researchers looked at children’s behaviors on the playground, observing that both typical children and children with SPD were likely to engage in participatory and regular games. In addition, they tend to play in small groups and spend less time with adults. However, there were some differences in terms of conflicts, play chance, and social signal awareness. In children with SPD, significant positive and negative correlations with the above items were found [20].

Articles on children with potential sensory processing disorder
Chien et al.’s studied children with potential sensory processing disorders, reporting that they had a significantly higher level of participation (in all subscales of the questionnaire and participation in child occupations) and less fun (in all areas except daycare) compared to children with normal sensory processing. The results of this study are as follows:
1. As for the level of participation, low sensory response and sensory seeking were the only predictors of the daycare area.
2. Low energy/weakness was the only significant predictor of participation pleasure.
3. As for the frequency of participation, low sensory response/sensory seeking, as well as the sensitivity of visual or auditory sensations, were two main predictors of participation in school activities.
4. Taste/olfactory sensitivity was not associated with participation in childhood activities.
5. There were no predictors in routine jobs and habits [21].
Articles on children with attention-deficit/hyperactivity disorder 
There was only one study on children with attention-deficit/hyperactivity disorder (ADHD), in which Engel-Yeger and Danila Ziv-On demonstrated a positive correlation between intensified hearing filter problems and low preference for social activities, as well as between leisure and informal activities and overall preference for all activities based on the Preference for Activities of Children (PAC) questionnaire. In addition, a significant correlation was observed between general sensory processing problems and a slight preference for participation in social activities [22].

Articles on children with atypical dermatitis
In this group of research, Engel-Yeger et al. observed positive and significant correlations as follows:
1. Low sensory response and sensory seeking correlate with a desire to participate in leisure activities.
2. Sensitivity to visual/auditory sensations correlates with the desire to participate in leisure activities.
3. The total score of sensory processing based on the short sensory profile test correlates with the desire to participate in leisure activities [23].
Their results suggested that children who have a problem with patterns of sensory processing are more likely to engage in recreational activities.

Articles on typically developing children
Walters and Lawson identified a significant positive correlation between parental sensory seeking and responsive play. They also found a correlation between sensory sensitivity and sensory avoiding scores of parents and children [24].
In the study of Ismail et al., a significant association was reported between sensory avoidance and more visits to public libraries/studies, puzzle-making, and daily activities. In their study, watching TV and movies was the absolute favorite of all children, while fishing, gardening, horse riding, and school activities were their least favorite in all four sensory patterns. In addition, significant negative correlations were reported as follows:
1. Between low sensory registration and overall participation diversity scores of children’s assessment of participation and enjoyment (CAPE),
2. Between sensory sensitivity and desires for social and skill-based activities,
3. Between sensory avoidance and desires for social and skill-based activities [25].

Measurement of sensory processing and participation in leisure and play 
Ten studies used the sensory profile test (the short or long form), one used the sensory experiences questionnaire (third edition), and one employed the sensory processing measure to assess sensory processing.
The following tools and questionnaires were used to assess participation in games and leisure: Activities
1. Parent-child play scale (one article)
2. Children’s assessment of participation and enjoyment-CAPE (three articles)
3. Preference for activities of children- PAC (three articles)
4. Playground observation forms (one article)
5. Child behavior checklist: Competence scales (one article)
6. Home and community activities scale

Meta-analysis
In this study, 6 articles [15, 16, 17, 23, 24, 25] that met the inclusion criteria were subjected to the quantitative synthesis. The results suggested significant correlations between the energy level and preference for participation in leisure activities (r=0.368, 95% CI: 0.23-0.49, P<0.001, I2=0.0), the auditory/visual sensitivity and preference for participation in leisure activities (r=0.505, 95% CI: 0.33-0.64, P<0.001, I2=0.0), the sensory processing patterns and the intensity of participation in leisure activities (r=0.109, 95% CI: 0.04-0.17, P=0.001, I2=0.50), the general sensory processing and preference for participation in leisure activities (r=494, 95% CI: 0.24-0.68, P<0.001, I2=0.63), the sensory processing patterns and preference for participation in informal activities (r=0.556, 95% CI: 0.32-0.73, P<0.001, I2=0.09), the olfactory sensitivity and preference for participation in leisure activities (r=0.527, 95%CI: 0.24-0.73, P=0.001, I2=0.0), the auditory filtering and preference for participation in leisure activities (r=0.623, 95% CI: 0.45-0.75, P<0.001, I2=0.51), the sensory processing patterns and enjoyment of participation in leisure activities (r=0.100, 95%CI: 0.02-0.18, P=0.013, I2=0.36), the sensory processing patterns and preference for participation in recreational activities (r=0.296, 95% CI: 0.14-0.43, P<0.001, I2=0.69). No significant correlations were found between the sensory sensitivity and desire to participate in social activities, the taste/olfactory sensitivity and intensity of engagement in leisure activities, the sensory sensitivity and “with whom” dimension of participation in leisure activities, the sensory seeking and preference for participation in leisure activities, the taste/olfactory sensitivity and delight of the involvement in leisure activities, the sensory sensitivity and “where” dimension of engagement in leisure activities, the sensory seeking and intensity of participation in leisure activities (Figures 2 to 17).

Publication bias 
Six studies were summarized in a funnel plot (Figure 18).

No publication bias was found in the meta-analysis based on Begg and Mazumdar’s correlation.

4. Discussion
As mentioned, the evidence presented by articles on the patterns of sensory processing and participation in play and leisure activities is inadequate. In the reviewed studies, leisure activities were classified into 7 categories: recreational, social, physical, self-improvement, skill-based, informal, and formal. Moreover, in all reviewed articles, a significant correlation had been reported between the patterns of sensory processing and childhood activity participation, though the direction and type of these correlations were different. However, a significant correlation emerged between general sensory processing problems and lesser involvement in leisure activities in three groups of studies, including children with ASD, potential sensory processing disorder, and ADHD [18, 21, 22]. Also, 9 out of 12 articles directly investigated the correlation between sensory processing patterns and play or leisure activities [3, 14, 24, 25]. Two out of the other three articles investigated the correlation between these patterns and participation in all children’s activities [18, 21], and one looked at their relationships with engagement in home and community activities [17]. Two articles consistently asserted that autistic children with strong sensory seeking are less likely to participate in outdoor activities than other activities [15, 17].
The meta-analysis of six studies [15, 16, 17, 23, 24, 25] illustrated significant correlations between the energy level and preference for participation in leisure activities, the auditory/visual sensitivity and preference for participation in leisure activities, the sensory processing patterns and the intensity of the involvement in leisure activities, the general sensory processing and preference for participation in leisure activities, the sensory processing patterns and preference for participation in informal activities, the olfactory sensitivity and preference for participation in leisure activities, the auditory filtering and preference for participation in leisure activities, the sensory processing patterns and enjoyment of participation in leisure activities, sensory processing patterns and preference for participation in recreational activities. 
In one of the studies, sensory avoidance was found to be significantly correlated with preferences for participation and intensity of involvement in leisure activities [25]. Therefore, it was impossible to carry out a meta-analysis, and there was scant evidence demonstrating the relationship between sensory avoidance patterns and leisure participation dimensions. 
The relationship between auditory/visual sensitivity and the preference for participation in leisure activities has been investigated in three studies in connection to four activity categories, including recreational, informal, social, and self-improvement activities [15, 22, 23], with the results indicating the significance of this relationship. However, no study has explored the relationship between auditory/visual sensitivity and other leisure activities. In addition, the association between general sensory processing and preference for participation in leisure activities had been examined for social [22] and recreational [23] activities, but other types of leisure activities have not been explored in studies. Although three studies had reported a significant relationship between sensory seeking and preferences for participating in physical [22]. self-improvement [15], and recreational activities [23], one study demonstrated a non-significant relationship between sensory seeking and preferences for participating in recreational activities [25]. The meta-analysis of these activities came together because they were part of leisure activities. Finally, the meta-analysis of two studies manifested a significant relationship between sensory processing patterns and preferences for participation in informal activities [15, 22]. These relationships were obtained for auditory filtering, auditory/visual, and motor sensitivity.
As far as the participation dimension is concerned, the relationship between taste/olfactory sensitivity and dimensions of participation in leisure activities was examined only in terms of the intensity and enjoyment of participation [15, 21]. Furthermore, among the sensory processing patterns, the association between sensory sensitivity and sensory seeking was investigated in terms of the dimensions of participation in leisure activities (diversity, intensity, with whom, where, enjoyment, frequency, and level) [15, 21, 25]. The results of the meta-analysis showed no significant relationship between the two activities. Since the relationship of sensory avoiding and sensory registration with the dimensions of leisure participation was only explored in Ismail et al.’s study [25], it could not be subjected to meta-analysis.
This study provided useful information on how different types of sensory processing patterns relate to dimensions of leisure and play participation in children. Knowledge of these relationships can lead to introducing new methods in the intervention process for children with disabilities.

5. Conclusion
The only conclusion that can be drawn from all patterns of sensory processing is the significant relationship between these patterns and the preference for participation in recreational activities (Figure 17). Among the patterns of sensory processing, studies on the relationship between sensory avoidance and the dimensions of leisure participation are fewer than the other patterns, and only one study mentioned it. Therefore, it is recommended that future studies focus more on this area to enable future meta-analyzes. The reviewed studies examined the relationships between sensory processing patterns and engagement in some types of leisure activities, and it is recommended that these relationships be examined with other kinds of activities as well. The association between the patterns of sensory processing and engagement in play and leisure activities has been studied in a limited clinical community, and they are very rare in the Middle East; therefore, these studies must be reviewed in children with a variety of disorders and be carried out in the Middle East. Finally, it should be noted that due to the low to moderate level of the available evidence, further research is needed. However, this study provides an understanding of the correlation between different patterns of sensory processing and play, and leisure activity participation based on scientific documents and demonstrates the significance of this concept in the occupational progress of children.

Limitations 
This review only covered articles published in English and Persian whose full text was available. Furthermore, participation in play and leisure activities encompassed playing with toys, unforced and active engagement of the children, and an internally-driven non-committed activity. It is performed voluntarily when not assigned to mandatory occupations such as sleep, work, or self-care. However, various definitions of play and leisure can change the inclusion criteria for future studies.

Ethical Considerations
Compliance with ethical guidelines
The Iran National Committee approved this research for Ethics in Biomedical Research (Code: IR.SBMU.RETECH.REC.1399.1344).

Funding
This research did not receive any grant from funding agencies in the public, commercial, or non-profit sectors. 

Authors' contributions
All authors equally contributed to preparing this article.

Conflict of interest
The authors declared no conflict of interest.

Acknowledgments
We would like to express our gratitude to everyone who contributed to the data collection and writing of this systematic review and meta-analysis. 

References

1. Moore A, Lynch H. Play and play occupation: A survey of paediatric occupational therapy practice in Ireland. Irish Journal of Occupational Therapy. 2018; 46(1):59-72. [DOI:10.1108/IJOT-08-2017-0022]
2. Case-Smith J, O'Brien JC. Occupational therapy for children-E-Book. Amsterdam: Elsevier Health Sciences; 2013. [Link]
3. Watts T, Stagnitti K, Brown T. Relationship between play and sensory processing: A systematic review. American Journal of Occupational Therapy. 2014; 68(2):e37-46. [DOI:10.5014/ajot.2014.009787] [PMID]
4. Dunn W. Supporting children to participate successfully in everyday life by using sensory processing knowledge. Infants & Young Children. 2007; 20(2):84-101. [DOI:10.1097/01.IYC.0000264477.05076.5d]
5. Bundy AC, Shia S, Qi L, Miller LJ. How does sensory processing dysfunction affect play? American Journal of Occupational Therapy. 2007; 61(2):201-8. [DOI:10.5014/ajot.61.2.201] [PMID]
6. Lawson LM, Dunn W. Children’s sensory processing patterns and play preferences. Annual in Therapeutic Recreation. 2008; 16(1). [Link]
7. Stagnitti K. Understanding play: The implications for play assessment. Australian Occupational Therapy Journal. 2004; 51(1):3-12. [DOI:10.1046/j.1440-1630.2003.00387.x]
8. Reilly M. Play as exploratory learning: Studies of curiosity behavior: California: SAGE Publications, Incorporated; 1974.
9. American Occupational Therapy Association. Occupational therapy practice framework: Domain and process. American Journal of Occupational Therapy. 2017; 68(Supplement_1):S1-48. [DOI:10.5014/ajot.2014.682006]
10. Hofferth SL, Sandberg JF. How American children spend their time. Journal of Marriage and Family. 2001; 63(2):295-308. [DOI:10.1111/j.1741-3737.2001.00295.x]
11. World Health Organization. International classification of functioning, disability, and health: Children & youth version: ICF-CY. Geneva: World Health Organization; 2007. [Link]
12. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Medicine. 2009; 6(7):e1000097. [DOI:10.1371/journal.pmed.1000097] [PMID] [PMCID]
13. Mostafaei H, Sadeghi-Bazargani H, Hajebrahimi S, Salehi-Pourmehr H, Ghojazadeh M, Onur R, et al. Prevalence of female urinary incontinence in the developing world: A systematic review and meta-analysis-A Report from the Developing World Committee of the International Continence Society and Iranian Research Center for Evidence Based Medicine. Neurourology and Urodynamics. 2020; 39(4):1063-86. [DOI:10.1002/nau.24342] [PMID]
14. Kuhaneck HM, Britner PA. A preliminary investigation of the relationship between sensory processing and social play in autism spectrum disorder. OTJR: Occupation, Participation and Health. 2013; 33(3):159-67. [DOI:10.3928/15394492-20130614-04] [PMID]
15. Hochhauser M, Engel-Yeger B. Sensory processing abilities and their relation to participation in leisure activities among children with high-functioning autism spectrum disorder (HFASD). Research in Autism Spectrum Disorders. 2010; 4(4):746-54. [DOI:10.1016/j.rasd.2010.01.015]
16. Reynolds S, Bendixen RM, Lawrence T, Lane SJ. A pilot study examining activity participation, sensory responsiveness, and competence in children with high functioning autism spectrum disorder. Journal of Autism and Developmental Disorders. 2011; 41(11):1496-506. [DOI:10.1007/s10803-010-1173-x] [PMID] [PMCID]
17. Little LM, Ausderau K, Sideris J, Baranek GT. Activity participation and sensory features among children with autism spectrum disorders. Journal of Autism and Developmental Disorders. 2015; 45(9):2981-90. [DOI:10.1007/s10803-015-2460-3] [PMID] [PMCID]
18. Loh SY, Ee SI, Marret MJ, Chinna K. Sensory processing and participation in childhood occupation in Autism and Typically developing children-A cross sectional case control. bioRxiv. 2020; 1-16. [Preprint]. [DOI:10.1101/2020.07.22.215574]
19. Cosbey J, Johnston SS, Dunn ML. Sensory processing disorders and social participation. American Journal of Occupational Therapy. 2010; 64(3):462-73. [DOI:10.5014/ajot.2010.09076] [PMID]
20. Cosbey J, Johnston SS, Dunn ML, Bauman M. Playground behaviors of children with and without sensory processing disorders. OTJR: Occupation, Participation and Health. 2012; 32(2):39-47. [DOI:10.3928/15394492-20110930-01]
21. Chien CW, Rodger S, Copley J, Branjerdporn G, Taggart C. Sensory processing and its relationship with children’s daily life participation. Physical & Occupational Therapy in Pediatrics. 2016; 36(1):73-87. [DOI:10.3109/01942638.2015.1040573] [PMID]
22. Engel-Yeger B, Ziv-On D. The relationship between sensory processing difficulties and leisure activity preference of children with different types of ADHD. Research in Developmental Disabilities. 2011; 32(3):1154-62. [DOI:10.1016/j.ridd.2011.01.008] [PMID]
23. Engel-Yeger B, Shani-Adir A, Kessel A. Participation in leisure activities and sensory modulation deficiencies of children with atopic dermatitis. Acta Paediatrica. 2011; 100(10):e152-e7. [DOI:10.1111/j.1651-2227.2011.02328.x] [PMID]
24. Welters-Davis M, Mische Lawson L. The relationship between sensory processing and parent-child play preferences. Journal of Occupational Therapy, Schools, & Early Intervention. 2011; 4(2):108-20. [DOI:10.1080/19411243.2011.595300]
25. Ismael NT, Lawson LAM, Cox JA. The relationship between children’s sensory processing patterns and their leisure preferences and participation patterns. Canadian Journal of Occupational Therapy. 2015; 82(5):316-24. [DOI:10.1177/0008417415577421] [PMID]