A systematic review of how adherence to caregiver facilitated therapeutic activities for children are assessed and reported in published research

Abstract

This systematic review assessed four measurement properties of measures (instruments) used by researchers to assess adherence to caregiver facilitated therapeutic activities (CFTA) for children in published research. The measurement properties considered in this systematic review were dyadic considerations, interpretability, reliability, and validity of the instruments used to assess adherence or compliance. Two databases (http://www.pubmed.gov and EMBASE) were searched for studies that reported adherence or compliance to CFTA prescribed by physiotherapists or occupational therapists or speech and language therapists. Papers included in this review were those that studied children less than 18 years old and or their primary caregivers. Data were extracted by the authors using a data extraction table adapted from the work of Bollen and colleagues (2014) and by rating the measurement properties of the adherence measures identified based on predefined rating criteria (see supplementary file and Table 1 respectively). The authors relied only on the published research articles and any associated published supplementary files/data. None of the 40 adherence instruments identified wholly fulfilled all the criteria of the four measurement properties assessed. The results of this systematic review show that the measures used by researchers for assessing adherence to CFTA generally lack validity, reliability, and dyadic considerations.

Keywords

Adherence compliance dyadic considerations interpretability reliability validity

1 Introduction

In this paper, compliance and adherence are used interchangeably.

Adherence or compliance to prescribed therapeutic activity is the accurate performance of the activity in line with the prescribed frequency, duration, and quality of performance [1, 2]. In a systematic review by Bollen and colleagues [3], 61 measures for assessing self-reported adherence to unsupervised home-based exercise were examined for their psychometric properties. The authors found that only eight of the measures examined had positive to intermediate content validity, three had intermediate reliability, while none had positive interpretability. The work of Bollen and colleagues exposed the inadequacy of measures of assessing adherence that are commonly used in clinical research [3]. Research in adherence to caregiver facilitated therapeutic activity (CFTA) in the paediatric population has received significant attention recently but the measurement properties of the measures of adherence (methods/instrument for assessing adherence) to CFTA have received less attention.

Adherence requires behaviour or lifestyle changes corresponding with recommendations agreed upon with a healthcare provider [4]. The paediatric population receiving rehabilitation services are generally involuntary patients brought to therapy by parents or caregivers. Oftentimes, the rehab process extends beyond the clinic where the caregiver is required to facilitate the performance of therapeutic activities for the child or with them in the absence of the therapist. Such CFTA require behavioural or lifestyle changes by both the caregiver and the child. Therefore, measures of adherence to CFTA should consider the dyad (caregiver and child) and not just the facilitator or caregiver. Studies that investigated adherence to CFTA for children tend to focus on the caregiver and less on the child [5, 6]. An important phenomenon that should prompt the inclusion of the child in the assessment of adherence to CFTA is the psychological concept of health locus of control, which is the belief regarding the type or amount of control that an individual has on their personal health based on personal experience(s) [7]. The health locus of control in an individual with a health condition could be internal, meaning that they believe that they have control over their own health and are personally responsible for it. Conversely, a person with an external health locus of control believes that other people or factors around them are responsible for their health. This person believes that they have no control over their own health and are therefore not responsible for it. It is also possible that the health locus of control in one individual may be internal for some health issues and external for others. Health locus of control may affect the health behaviours of children, including compliance to prescribed therapeutic activity [8 –10]. Therefore, it is imperative that methods of assessing adherence to CFTA should assess dyadic adherence instead of focusing solely on the adherence of the caregiver.

A caregiver is anyone who provides care for another person with a health condition. Caregivers may be family members, friends, or anyone within the social support network of persons with health conditions [11]. Their caregiving duties are voluntary and uncompensated. Thus, they are often referred to as informal caregivers as opposed to paid caregivers who are employed to lessen the caregiving burden of the family members. Given that the caregiving tasks of paid caregivers are considered routine activities within their job description, they were excluded from this review in order to focus on informal caregivers.

2 Study aims

This study aimed to identify measures of adherence to CFTA for children that have been used in published research and to assess the way published research papers report the measurement properties of the instruments they used to assess adherence. The CFTA considered in this study are those prescribed by physiotherapists, occupational therapists, and speech and language therapists. Also, the study aimed to document the existence or lack of four measurement properties in the identified measures of adherence. The first measurement property considered was content or construct validity expressed statistically by Pearson’s r value and its equivalents. This is the strength of the correlation between measurements obtained from an instrument compared to measurements obtained from a referent standard instrument that measures the same construct [12 –14]. The second measurement property was reliability (inter-rater, or test-retest reliability, or internal consistency reliability) [13, 14]. This is the level of agreement between the measurement values obtained from repeated use of a measuring instrument either by one assessor or multiple assessors. It is also determinable by Pearson’s r value and its equivalents. The third property considered was interpretability which may be defined as either a clear rationale for determining meaningful changes in adherence, or a clear rationale for making differences in adherence between participants or changes in the level of adherence for individual participants over time meaningful [12]. Lastly, dyadic consideration has been considered as a dichotomous characteristic of adherence measures in this systematic review. This classified the measures of adherence as either with dyadic consideration (adherence of both the caregiver and the child), or without dyadic consideration (adherence of either the caregiver alone or the child alone) [15].

3 Methods

Two databases (http://www.pubmed.gov and EMBASE) were searched on the 11^th of December 2020 for the following combination of keywords/key phrases: ((caregiver or family) AND (compliance or adherence)) AND (physiotherapy or occupational therapy or speech and language therapy). Broad search terms were used in order to retrieve the maximum number of relevant publications. The search results were then filtered to include human species and articles that have been written in the English language. The study selection process is shown in Fig. 1.

Fig. 1

PRISMA Flow chart of the search strategy and study selection.

Papers included in this review fulfilled the following criteria: the study participants (population) must have been unpaid or informal caregivers (parents, family members, friends, guardians, etc) providing care for children (aged from birth to 18 years old) and/or the children receiving care from them. The CFTA (intervention) were prescribed by any of the three professionals stated in the study aim. An exception to this inclusion criterion was the inclusion of studies that investigated adherence to foot abduction braces as part of the treatment for congenital talipes equinovarus. This exception was necessary because the prescription and monitoring of foot abduction braces is done by either orthopaedic surgeons [16] or physiotherapists [17] depending on the region. Furthermore, the interventions considered in this review included passive therapeutic activities performed by the caregiver. There were no exclusions based on disease conditions or the types of medical interventions or surgical interventions associated with the health conditions of the children that participated in the studies selected. The outcomes of interest in this review are the measurement properties of the instruments/methods used to assess adherence to CFTA. Therefore, to be included in this review a study must have clearly described the method (or instrument) used to assess adherence. Studies that did not assess adherence to CFTA were excluded. Regarding the study type, the authors included any type of completed study in which adherence to CFTA for children was assessed. Protocols of studies not yet completed were excluded.

The two authors independently applied the study selection criteria to the studies retrieved from both databases searched and, by consensus, decided on the inclusion or exclusion of the ones on which they did not initially agree. Data was extracted from the full text of each paper selected for inclusion by using a data extraction table adapted from the work of Bollen and colleagues on the psychometric properties of self-reported adherence to unsupervised home-based exercise [3]. See supplementary file for the data extraction table.

The full text of one paper published in 1993 was only accessible via a paywall. It was excluded from this review because it was not available through the authors’ institutional access for published research.

4 Rating of measurement properties of identified measures of adherence

The measurement properties of the identified adherence measures were rated according to the criteria described in Table 1. For studies that did not report validity statistics, it was possible to assess the validity of the adherence instrument based on the criteria and rating system for evaluating the content validity of patient reported outcome measure developed by Consensus Based Standards for the Selection of Health Measurement Instruments (COSMIN) [18]. However, in order to do this the adherence measure must have been either adequately described in the reviewed paper or the details of the adherence measure must have been available in associated supplementary data or the paper made reference to another publication where the adherence measure was adequately described. The criteria for rating the measurement properties described in Table 1 are for descriptive purposes only. Based on these criteria, the maximum rating score for validity was 2. Similarly, the maximum score for reliability was 2. Dyad and interpretability each had a maximum rating score of 1. Therefore, for each adherence measure, the maximum overall score for all four measurement properties combined was 6.

Table 1
Criteria for rating the measurement properties of methods of assessing adherence to caregiver facilitated therapeutic activity

Validity Content or construct Rate 2 points if Validity is reported in the paper as tested and the r value or equivalent is≥0.7 or≥ + 7/10 on COSMIN^$. Rate 1 point if Validity is reported in the paper as tested and the r value or equivalent is≥0.55 but < 0.7 or ≥+ 5/10 but < + 7/10 on COSMIN^$. Rate 0 points if Validity is reported in the paper as tested and the r value or equivalent is < 0.55 or < 5/10 on COSMIN^$or the paper did not report validity testing and there is no reference to another study reporting validity testing.

Reliability Test-retest reliability or inter-rater reliability or internal consistency reliability Rate 2 points if Reliability is reported in the paper as tested and the r value or equivalent is≥0.7^. Rate 1 point if Reliability is reported in the paper as tested and the r value or equivalent is≥0.55 but < 0.7^. Rate 0 points if Reliability is reported in the paper as tested and the r value or equivalent is < 0.5 or the paper did not report reliability testing and there is no reference to another study reporting reliability testing.

Interpretability Responsiveness to change Rate 1 point if Interpretability is clearly and meaningfully described in the study or there is reference to another publication in which interpretability is clearly described or interpretability is inherently clear from the measure of adherence, e.g., units of time or a scale, etc. Rate 0 points if The paper did not clearly and meaningfully describe interpretability and there is no reference to another study in which interpretability is clearly described and Interpretability is not inherently clear from the measure of adherence.

Dyad Adherence measure incorporates dyadic considerations Rate 1 point if The adherence measure considers both the caregiver and the child. Rate 0 points if The adherence measure considers either the caregiver or the child but not both.

r value = Intraclass Correlation Coefficient [34]; Equivalents of r = Cohen’s Kappa, % agreement; ^*Frost et al (2007) [34]; ^$COSMIN criteria and rating system for evaluating the content validity of Patient Reported Outcome Measure [18]. Equivalent of r value for reliability = Cronbach’s Alpha.

Considering that adherence to CFTA is not a clinical outcome but an intermediate factor required for a desired clinical outcome [4, 19], the measurement property of interpretability in Table 1 is not based on the responsiveness of adherence measures to detect clinically important change. Furthermore, the concept of “minimum important change” or its equivalents are not considered in the context of this systematic review. Rather, the rating of interpretability of adherence measures in this study relates to the capacity of each adherence measure to detect any change (not necessarily clinically relevant change) in adherence levels within one participant or the differences between participants. This is assessed based on the clarity and relevance of the definition and or the description of interpretability by the creators of the measures of adherence [20]. The rating of interpretability is also based on the relevance of each adherence measure to the population for which it was used. Some adherence measures are inherently interpretable because they have numerical scales, Likert scales or similar reporting tools. Such adherent measures were automatically given a score of 1 for interpretability in this review (see Table 1). Although it is important for researchers of adherence to state the threshold or minimum score that they categorised as adherent in their instrument, this was not considered in the rating of interpretability in this review.

5 Results

Twenty-nine papers published between 1981 and 2020 were selected for inclusion in this systematic review. They featured diverse methods for assessing CFTA adherence which were grouped into 7 categories shown in Table 2. Some of the papers featured more than one method of assessing adherence to CFTA. A total of 40 adherence measures were identified from the 29 papers reviewed. Details of the data extraction are in the supplementary file.

Table 2
Measures (instruments) of adherence to caregiver facilitated therapeutic activity

Category of methods of assessing adherence to CFTA Category abbreviated name Description Number of publications that used the method

Activity log Log Diary, activity logbook, tally sheet, online activity log, paper log, and mobile phone app log. 15

Disposable treatment item Disposable The number of unused disposable treatment items. 2

Observation Observation Video recording of CFTA viewed by researcher and in-person observation of performance of CFTA by researcher. 2

Questionnaire and interview Quest Self-reported questionnaire (including online), therapist administered questionnaire, researcher administered questionnaire, interviews (on phone or in person), parent self-report in person or retrieved from medical record, and standardised forms. 18

Sensor built into therapeutic devices Sensor Chronometer built into devices such as chest compression vests. 1

Therapist’s estimate Estimate. Subjective estimation of therapist. 1

Visual Analogue scale VAS Visual analogue scale. 1

Category of methods of assessing adherence to CFTA	Category abbreviated name	Description	Number of publications that used the method
Activity log	Log	Diary, activity logbook, tally sheet, online activity log, paper log, and mobile phone app log.	15
Disposable treatment item	Disposable	The number of unused disposable treatment items.	2
Observation	Observation	Video recording of CFTA viewed by researcher and in-person observation of performance of CFTA by researcher.	2
Questionnaire and interview	Quest	Self-reported questionnaire (including online), therapist administered questionnaire, researcher administered questionnaire, interviews (on phone or in person), parent self-report in person or retrieved from medical record, and standardised forms.	18
Sensor built into therapeutic devices	Sensor	Chronometer built into devices such as chest compression vests.	1
Therapist’s estimate	Estimate.	Subjective estimation of therapist.	1
Visual Analogue scale	VAS	Visual analogue scale.	1

Table 3 shows that none of the 40 measures of adherence identified achieved the maximum rating of 6/6 for the four measurement properties considered.

Table 3

Ratings of the measurement properties of adherence measures based on the criteria in Table 1

#	Citation	Adherence measure	Validity rating (/2)	Dyad rating (/1)	Interpretability rating (/1)	Reliability rating (/2)	Total (/6)
1	Frisch et al., 2020 [27]	Quest	0	1	1	0	2
2		Sensor	0	0	1	0	1
3	Oates et al., 2019 [35]	Quest	0	0	1	2	3
4	Limpaphayom and Sailohit, 2019 [36]	Quest	0	0	0	0	0
5	Charpentier et al., 2020 [37]	Observation	0	0	1	1	2
6		Log	0	1	1	0	2
7	Christmas et al., 2018 [26]	Quest	0	1	1	0	2
8	Medina-Mirapeix et al., 2017 [38]	Quest	0	0	1	0	1
9	Ferre et al., 2015 [25]	Quest	0	1	1	0	2
10	Göksan et al., 2015 [28]	Quest	0	1	0	0	1
11	Cruz et al., 2014 [39]	Quest	0	0	1	0	1
12	McQueen et al., 2013 [40]	Observation	0	0	0	0	0
13		Log	0	0	1	0	1
14	Rabino et al., 2013 [41]	Log	0	0	1	0	1
15	Longmuir et al., 2013 [42]	VAS	0	0	1	0	1
16	Carr Swift et al., 2011 [43]	Log	0	0	1	0	1
17	Hung &Pang, 2010 [44]	Log	0	0	1	0	1
18		Log	0	0	0	0	0
19	Gelfer et al., 2010 [45]	Quest	0	0	0	0	0
20	Yuan et al., 2010 [46]	Quest	0	0	0	0	0
21	Rivett et al., 2009 [47]	Log	0	0	1	0	1
22	Downs et al., 2006 [29]	Log	0	0	0	1	1
23	Charles et al., 2006 [48]	Log	0	0	0	0	0
24		Quest (GAS)	0	0	1	2	3
25	De Civita et al., 2005 [5]	Quest (PARQ)	1	1	1	1	4
26		Log	0	0	1	0	1
27	Dobbs et al., 2004 [49]	Quest	0	0	0	0	0
28	Galil et al., 2001 [6]	Quest	0	0	1	0	1
29	Steinbok et al., 1997 [50]	Log	0	0	1	0	1
30	Schreiber et al., 1995 [51]	Log	0	0	1	0	1
31		Quest	2	0	1	2	5
32	Law &King, 1993 [21]	Log	0	0	1	0	1
33		Disposable	0	0	1	0	1
34	Gajdosik &Campbell, 1991 [52]	Log	0	0	1	0	1
35		Estimate	0	0	0	0	0
36		Quest	0	0	0	0	0
37	Mayo, 1981 [53]	Log	0	0	1	0	1
38		Disposable	0	0	1	0	1
39	Johnson et al., 1981 [54]	Quest	0	0	0	0	0
40	Marshall &Sheppard, 1981 [55]	Quest	0	0	0	0	0

PARQ (Parent Adherence Report Questionnaire). GAS (General Adherence Scale).

Only one adherence measure achieved a total rating of 5/6. This was the compliance questionnaire used by Law and King [21]. It fell short of the maximum rating because it lacked dyadic considerations. The next highest rating was the Parent Adherence Report Questionnaire used by De Civita and colleagues [5]. It scored 4/6 and is the only adherence measure that wholly or partially fulfilled all the 4 measurement properties considered (i.e. validity, dyadic consideration, interpretability and reliability). It was rated 1/2 for construct validity because the reported r value was 0.62 when the General Adherence Scale is used as the referent standard scale that measured the same construct. This means that there was a 62% agreement between the two. De Civita and colleagues also reported a Kappa coefficient of 0.36 which they computed by comparing the Parent Adherence Report Questionnaire to adherence rates reported in the daily logs of caregivers in their study. This means that there was 36% agreement between values obtained from the Parent Adherence Report Questionnaire and the caregivers’ reported adherence in their daily logs of CFTA performed. Similarly, the reliability (reproducibility) of the exercise items on the Parent Adherence Report Questionnaire was rated 1/2 in this review because the intraclass correlation coefficient for reproducibility was reported to range from 0.53 to 0.88. This indicates that there was between 53% to 88% agreement between values obtained from repeated assessments for some of the items on the questionnaire. Furthermore, the Kappa coefficient for the reproducibility of items on the Parent Adherence Report Questionnaire assessing the locus of responsibility was 0.42. Therefore, the authors of this review downgraded the rating for reliability to 1/2 because the reliability coefficients of several items were less than 0.7.

Interpretability was the most commonly fulfilled measurement property. A total of 27 of the 40 adherence measures identified received a rating of 1/1 for interpretability. This was attributed to the way interpretability was defined in the rating criteria for the measurement properties (Table 1).

Only two of the 40 adherence measures reviewed reported validity data, six reported reliability and six had dyadic considerations. Twenty of the adherence measures reviewed had total scores of 1/6 while 11 of them had total ratings of 0/6. This indicated either a lack of consideration for any of the four measurement properties reviewed or inadequate reporting of the measurement properties by the authors of those research articles.

6 Discussion

The findings of this review are in line with those of Bollen and colleagues [3]. Researchers frequently use measures for assessing adherence with inadequate consideration for the measurement properties of the instruments used. This is perhaps attributable to the aforementioned fact that adherence is not regarded as a clinical outcome but rather an intermediate requirement for the achievement of a desired clinical outcome. Hence, the measurement properties of the instruments used to assess adherence tend to receive less scrutiny from research publication reviewers. However, this review also demonstrates the trend for inaccurate assessment of adherence by researchers and clinicians alike. The literature is replete with myriad and disparate methods of assessing adherence to CFTA which makes the consideration and accurate reporting of their measurement properties crucial to evidence-based practice. For example, Frost and colleagues in 2016 [22] reported that the use of diaries to assess adherence is valid at the group level but not at the individual level. Conversely, Frost and colleagues in 2017 [23] reported that diaries and logs have excellent criterion validity for the specific populations in which they have been validated for but lack reliability. Furthermore, observational scales for assessing adherence can be valid and reliable [24] but the researchers using them should demonstrate the validity and reliability of the specific scales used. The use of some wearable sensors for assessing adherence may be reliable but has inadequate validity [23]. The incoherence and divergence in the outcomes of these studies demonstrate that the validation of one adherence measure does not confer validity or reliability onto another adherence measure in the same category. It is vitally important that researchers should report the measurement properties of the specific instrument used in their work.

It is desirable for adherence measures for paediatric CFTA to have dyadic considerations because they seek to measure changes in the behaviours of both the child and the caregiver, rather than the behaviour of either the caregiver alone or the child alone. Studies that used adherence measures with dyadic considerations have included research participants as young as 9 months old [25]. This is because adherence to CFTA is usually a patient reported outcome (excepting some wearable sensors) and the caregiver is the reporter of child adherence to CFTA [26 –28]. Therefore, it follows that dyadic consideration is not limited by the age of the child, verbal ability, or cognitive ability. It is feasible to assess the adherence of the dyad in any CFTA. However, this review shows that adherence measures for paediatric CFTA generally lack dyadic considerations, which may necessitate the combination of two or more complementary adherence measures in the absence of a single measure that considers both the caregiver and the child. This is a viable solution for dyadic considerations in paediatric adherence assessment in clinical and research settings. Furthermore, future research should incorporate a dyadic framework into the development of new adherence measures for CFTA. This is important given an increasing global burden of diseases and the need to facilitate maximum adherence to non-pharmacological and non-surgical therapies. Accurate quantification of adherence to CFTA may add value for money to healthcare services per unit of population by enhancing the economic factors in rehabilitation. This includes recovery time, the number of consultations with rehabilitation specialists before full recovery, injury recurrence, and readmission rates.

7 Study limitations

This review relied solely on published research articles found on two databases and their published supplementary data, where available. The reviewers did not contact the authors of any of the articles reviewed for more data/information. This may affect the rating of the measurement properties of some of the adherence measures. For example, a cursory look at the one-week diary card used by Downs and associates [29] appeared to be inherently interpretable, but it was not available for the authors of this systematic reviewer to assess. In addition, Downs and associates did not adequately discuss the interpretability of their adherence measure, nor did they provide a reference to published supplementary data on the subject. Hence, the interpretability of the one-week diary card was rated as zero (i.e., not interpretable).

Another limitation of this study is that it did not distinguish the components of adherence assessed by each adherence measure identified. Considering that adherence is defined as behavioural changes in line with prescribed frequency, duration, and quality of performance of therapeutic activity [4, 29], an adherence measure should assess adherence to the three components (frequency, duration, and quality/accuracy of performance) of the prescribed therapeutic activity. It is rare to find adherence measures that assess all three components. A cursory look at the 40 adherence measures identified in the 29 papers reviewed showed that most of them did not. This could be due to the fact that most adherence measures are caregiver reported, and the frequency and duration of performance of CFTA are easily documentable in a diary or questionnaire. On the other hand, the accuracy of performance of CFTA requires observation by an expert performer (clinician) or a mechanism for detecting the accuracy of performance. This is perhaps the reason why most adherence measures are unable to assess all components of adherence. Future studies may explore the ability of measures to assess all components of adherence.

Finally, studies that reported the use of an adherence measure created by modifying an existing validated measure were rated for their measurement properties by the authors of this review based on the rating of the original adherence measure from which the modified measure was created. An example of such is the modified general adherence scale used by De Civita and associates [5] which was rated 2/2 for both validity and reliability because the original scale satisfied the criteria for those ratings. This rating process may not be accurate given that any modifications to a measurement scale requires revalidation of the scale [30 –32].

8 Conclusions

Most adherence measures used for paediatric CFTA are interpretable, but they are generally not valid or reliable. They also do not consider the dyad and, instead, focus mostly on the caregiver. Given that adherence to therapy is a requirement for desired clinical outcomes, the practical implication of this study is that, in the absence of evidence-based adherence monitoring strategies by clinicians and researchers, clinical effectiveness studies would be suboptimal, clinical governance would be incomplete, cost-benefit analysis of preventive and curative strategies in rehabilitation would be inaccurate, and there would remain a missing link in strategies to reduce the global burden of diseases [33]. It is proposed that researchers and clinicians use valid and reliable measures of adherence and that they clearly report the measurement properties of their adherence measures. Furthermore, in the absence of adherence measures for paediatric CFTA that have dyadic considerations, clinicians and researchers may combine two or more complementary adherence measures to ensure that the adherence of both the child and caregiver are assessed.

Footnotes

Acknowledgments

The authors acknowledge Roberto Meroni (PhD) for facilitating access to one of the databases used in this study.

Conflict of interest

The authors have no conflict of interest to report.

Ethical considerations

This paper did not require ethical review board approval because it is a systematic review of the literature.

The supplementary material is available in the electronic version of this article: .

References

Anar

SÖ

. The effectiveness of home-based exercise programs for low back pain patients. J Phys Ther Sci. 2016;28(10):2727–30. doi: 10.1589/jpts.28.2727

Jack

, McLean

, Moffett

, Gardiner

. Barriers to treatment adherence in physiotherapy outpatient clinics: a systematic review. Man Ther. 2010;15(3):220–8. doi: 10.1016/j.math.2009.12.004

Bollen

, Dean

, Siegert

, Howe

, Goodwin

. A systematic review of measures of self-reported adherence to unsupervised home-based rehabilitation exercise programmes, and their psychometric properties. BMJ Open. 2014;4(6):e005044. doi: 10.1136/bmjopen-2014-005044

World Health Organisation. Adherence to long-term therapies: evidence for action. 2003. Available from: https://apps.who.int/iris/bitstream/handle/10665/42682/9241545992.pdf

De Civita

, Dobkin

, Ehrmann-Feldman

, Karp

, Duffy

. Development and Preliminary Reproducibility and Validity of the Parent Adherence Report Questionnaire: A Measure of Adherence in Juvenile Idiopathic Arthritis. J Clin Psychol Med Settings. Mar. 2005;12(1):1–12. doi: 10.1007/s10880-005-0907-y

Galil

, Carmel

, Lubetzky

, Vered

, Heiman

. Compliance with home rehabilitation therapy by parents of children with disabilities in Jews and Bedouin in Israel. Dev Med Child Neurol. 2001;43(4):261–8. doi: 10.1017/s0012162201000494

Pourhoseinzadeh

, Gheibizadeh

, Moradikalboland

, Cheraghian

. The Relationship between Health Locus of Control and Health Behaviors in Emergency Medicine Personnel. Int J Community Based Nurs Midwifery.. 2017;5(4):397–407.

Perrin

, Shapiro

. Health locus of control beliefs of healthy children, children with a chronic physical illness, and their mothers. J Pediatr. 1985;107(4):627–33. doi: 10.1016/s0022-3476(85)80038-x

Nazareth

, Richards

, Javalkar

, Haberman

, Zhong

, Rak

, et al. Relating Health Locus of Control to Health Care Use, Adherence, and Transition Readiness Among Youths With Chronic Conditions, 2015. North Carolina, Prev Chronic Dis. 2016;13:E93. doi: 10.5888/pcd13.160046

10.

Travlos

, Drew

, Patman

. The value of the CoughAssist® in the daily lives of children with neuromuscular disorders: Experiences of families, children and physiotherapists. Dev Neurorehabil. 2016;19(5):321–6. doi: 10.3109/17518423.2014.993771

11.

Caregivers Nova Scotia. The Caregiver’s Handbook: Help for Unpaid Caregivers in Nova Scotia. Mar 2018. Available from: https://caregiversns.org/images/uploads/all/Caregivers_Handbook_April_2018_Web.pdf

12.

McClimans

. Interpretability, validity, and the minimum important difference. Theor Med Bioeth. 2011;32(6):389–401. doi: 10.1007/s11017-011-9186-9

13.

Karras

. Statistical Methodology: II. Reliability and Validity Assessment in Study Design, Part A. Acad Emerg Med. 1997;4(1):64–71. doi: 10.1111/j.1553-2712.1997.tb03646.x

14.

McHugh

. Interrater reliability: the kappa statistic. Biochem Med (Zagreb). 2012;22(3):276–82.

15.

Driscoll

, Schatschneider

, McGinnity

, Modi

. Application of dyadic data analysis in pediatric psychology: cystic fibrosis health-related quality of life and anxiety in child-caregiver dyads. J Pediatr Psychol. 2012;37(6):605–11. doi: 10.1093/jpepsy/jss063

16.

Besselaar

, Sakkers

RJB

, Schuppers

, Witbreuk

MMEH

, Zeegers

EVCM

, Visser

, et al. Guideline on the diagnosis and treatment of primary idiopathic clubfoot. Acta Orthop. 2017;88(3):305–9. doi: 10.1080/17453674.2017.1294416

17.

New South Wales Government. Agency for Clinical Innovation. Management of Infants and Children with Congenital Talipes Equinovarus. 2014. Available from: https://www1.health.nsw.gov.au/pds/ActivePDSDocuments/GL2014_014.pdf

18.

Terwee

, Prinsen

CAC

, Chiarotto

, de

Vet HCW

, Bouter

, Alonso

, et al. COSMIN methodology for assessing the content validity of PROMS. User manual version 1.0. 2018. Available from: https://cosmin.nl/wp-content/uploads/COSMIN-methodology-for-content-validity-usermanual-v1.pdf

19.

Hayden

, van Tulder

, Tomlinson

. Systematic review: strategies for using exercise therapy to improve outcomes in chronic low back pain. Ann Intern Med. 2005;142(9):776–85. doi: 10.7326/0003-4819-142-9-200505030-00014

20.

Louie

, D’Agostino

, Han

, Ryken

. Determining an Appropriate Outcome Measure in Neurosurgical Research: Investigating Meaningful, Valid, and Practical Metrics. Cureus. 2019;11(9):e5610. doi: 10.7759/cureus.5610

21.

Law

, King

. Parent compliance with therapeutic interventions for children with cerebral palsy. Dev Med Child Neurol. 1993;35(11):983–90. doi: 10.1111/j.1469-8749.1993.tb11580.x

22.

Frost

, McClurg

, Brady

, Williams

. Optimising the validity and completion of adherence diaries: a multiple case study and randomised crossover trial. Trials. 2016;17(1):489. doi: 10.1186/s13063-016-1615-7

23.

Frost

, Levati

, McClurg

, Brady

, Williams

. What Adherence Measures Should Be Used in Trials of Home-Based Rehabilitation Interventions? A Systematic Review of the Validity, Reliability, and Acceptability of Measures. Arch Phys Med Rehabil. 2017;98(6):1241–56.e45. doi: 10.1016/j.apmr.2016.08.482

24.

Arensman

, Geelen

, Koppenaal

, Veenhof

, Pisters

. Measuring exercise adherence in patients with lowback pain: development, validity, and reliability of the EXercise Adherence Scale (EXAS). Physiother Theory Pract. 2020; 1-10. doi: 10.1080/09593985.2020.1818337

25.

Ferre

, Brandão

, Hung

Y-C

, Carmel

, Gordon

. Feasibility of caregiver-directed home-based hand-arm bimanual intensive training: a brief report. Dev Neurorehabil. 2015;18(1):69–74. doi: 10.3109/17518423.2014.948641

26.

Christmas

, Sackley

, Feltham

, Cummins

. A randomized controlled trial to compare two methods of constraint-induced movement therapy to improve functional ability in the affected upper limb in pre-school children with hemiplegic cerebral palsy: CATCH TRIAL. Clin Rehabil. 2018;32(7):909–18. doi: 10.1177/0269215518763512

27.

Frisch

, Rosenblum

, Tirosh

. Parental Occupational Executive Training: Feasibility and Parental Perceptions. OTJR (Thorofare N J). 2020;40(3):203–10. doi: 10.1177/1539449220912191

28.

Göksan

, Bilgili

, Eren

, Bursalı

, Koç

. Factors affecting adherence with foot abduction orthosis following Ponseti method. Acta Orthop Traumatol Turc. 2015;49(6):620–6. doi: 10.3944/AOTT.2015.14.0348

29.

Downs

, Roberts

, Blackmore

, Le Souëf

, Jenkins

. Benefits of an education programme on the self-management of aerosol and airway clearance treatments for children with cystic fibrosis. Chron Respir Dis. 2006;3(1):19–27. doi: 10.1191/1479972306cd100oa

30.

Boateng

, Neilands

, Frongillo

, Melgar-Quiñonez

, Young

. Best Practices for Developing and Validating Scales for Health, Social, and Behavioral Research: A Primer. Front Public Health. 2018;6:149. doi: 10.3389/fpubh.2018.00149

31.

Stewart

, Thrasher

, Goldberg

, Shea

. A framework for understanding modifications to measures for diverse populations. J Aging Health. 2012;24(6):992–1017. doi: 10.1177/0898264312440321

32.

Yee

, Serrano

, Kando

, McElroy

. A psychometric analysis and revalidation of the Yale-Brown Obsessive Compulsive Scale modified for Binge Eating in adults with binge eating disorder. Qual Life Res. 2019;28(12):3385–94. doi: 10.1007/s11136-019-02277-8

33.

World Health Organization. Adherence to long-term therapies: Evidence for action. 2003. Available from: https://www.who.int/chp/knowledge/publications/adherencefull_report.pdf

34.

Frost

, Reeve

, Liepa

, Stauffer

, Hays

. What is sufficient evidence for the reliability and validity of patient-reported outcome measures? Value Health. 2007;10(Suppl 2):S94–S105. doi: 10.1111/j.1524-4733.2007.00272.x

35.

Oates

, Stepanikova

, Rowe

, Gamble

, Gutierrez

, Harris

. Objective Versus Self-Reported Adherence to Airway Clearance Therapy in Cystic Fibrosis. Respir Care. 2019;64(2):176–81. doi: 10.4187/respcare.06436

36.

Limpaphayom

, Sailohit . Factors Related to Early Recurrence of Idiopathic Clubfoot Post the Ponseti Method. Malays Orthop J. 2019;13(3):28–33. doi: 10.5704/MOJ.1911.005

37.

Charpentier

, Morgan

, Harding

. A service evaluation of parent adherence with dysphagia management therapy guidelines: reports from family carers supporting children with complex needs in Greece. Disabil Rehabil. 2020;42(3):426–33. doi: 10.1080/09638288.2018.1499048

38.

Medina-Mirapeix

, Lillo-Navarro

, Montilla-Herrador

, Gacto-Sánchez

, Franco-Sierra

MÁ

, Escolar-Reina

. Predictors of parents’ adherence to home exercise programs for children with developmental disabilities, regarding both exercise frequency and duration: a survey design. Eur J Phys Rehabil Med. 2017;53(4):545–55. doi: 10.23736/S1973-9087.17.04464-1

39.

Marchetti Cruz

, Aparecida Caromano

, Lopes Gonçalves

, Gaiad Machado

, Callil Voos

. Learning and adherence to baby massage after two teaching strategies. J Spec Pediatr Nurs. 2014;19(3):247–56. doi: 10.1111/jspn.12076

40.

McQueen

, Lakes

, Rich

, Vaughan

, Hayes

, Cooper

, et al. Feasibility of a caregiver-assisted exercise program for preterm infants. J Perinat Neonatal Nurs. 2013;27(2):184–92. doi: 10.1097/JPN.0b013e31828b244a

41.

Rarback Rabino

, Rokach Peretz

, Kastel-Deutch

, Tirosh

. Factors affecting parental adherence to an intervention program for congenital torticollis. Pediatr Phys Ther. Fall. 2013;25(3):298–303. doi: 10.1097/PEP.0b013e318298eb92

42.

Longmuir

, Tyrrell

, Corey

, Faulkner

, Russell

, McCrindle

. Home-based rehabilitation enhances daily physical activity and motor skill in children who have undergone the Fontan procedure. Pediatr Cardiol. 2013;34(5):1130–51. doi: 10.1007/s00246-012-0618-8

43.

Carr Swift

, O’Brian

, Hewat

, Onslow

, Packman

, Menzies

. Investigating parent delivery of the Lidcombe Program. Int J Speech Lang Pathol. 2011;13(4):308–16. doi: 10.3109/17549507.2011.550692

44.

Hung

WWY

, Pang

MYC

. Effects of group-based versus individual-based exercise training on motor performance in children with developmental coordination disorder: a randomized controlled study. J Rehabil Med. 2010;42(2):122–8. doi: 10.2340/16501977-0496

45.

Gelfer

, Durham

, Daly

, Shitrit

, Smorgick

, Ewins

. The effect of neuromuscular electrical stimulation on congenital talipes equinovarus following correction with the Ponseti method: a pilot study. J Pediatr Orthop B. 2010;19(5):390–5. doi: 10.1097/BPB.0b013e32833becad

46.

Yuan

, Kane

, Shelton

, Matel

, Becker

, Moss

. Safety, tolerability, and efficacy of high-frequency chest wall oscillation in pediatric patients with cerebral palsy and neuromuscular diseases: an exploratory randomized controlled trial. J Child Neurol. 2010;25(7):815–21. doi: 10.1177/0883073809350223

47.

Rivett

, Rothberg

, Stewart

, Berkowitz

. The relationship between quality of life and compliance to a brace protocol in adolescents with idiopathic scoliosis: a comparative study. BMC Musculoskelet Disord. 2009;10:5. doi: 10.1186/1471-2474-10-5

48.

Charles

, Wolf

, Schneider

, Gordon

. Efficacy of a child-friendly form of constraint-induced movement therapy in hemiplegic cerebral palsy: a randomized control trial. Dev Med Child Neurol. 2006;48(8):635–42. doi: 10.1017/S0012162206001356

49.

Dobbs

, Rudzki

, Purcell

, Walton

, Porter

, Gurnett

. Factors predictive of outcome after use of the Ponseti method for the treatment of idiopathic clubfeet. J Bone Joint Surg Am. 2004;86(1):22–7. doi: 10.2106/00004623-200401000-00005

50.

Steinbok

, Reiner

, Kestle

. Therapeutic electrical stimulation following selective posterior rhizotomy in children with spastic diplegic cerebral palsy: a randomized clinical trial. Dev Med Child Neurol. 1997;39(8):515–20. doi: 10.1111/j.1469-8749.1997.tb07479.x

51.

Schreiber

, Effgen

, Palisano

. Effectiveness of Parental Collaboration on Compliance with a Home Program. Pediatric Physical Therapy. 1995;7(2):59–64.

52.

Giller Gajdosik

, Campbell

. Effects of Weekly Review, Socioeconomic Status, and Maternal Belief on Mothers’ Compliance with Their Disabled Children’s Home Exercise Program. Physical & Occupational Therapy In Pediatrics. 1991;11(2):47–65. doi: 10.1080/J006v11n02_04

53.

Mayo

. The effect of a home visit on parental compliance with a home program. Phys Ther. 1981;61(1):27–32. doi: 10.1093/ptj/61.1.27

54.

Johnson

, Gershowitz

, Stabler

. Maternal compliance and children’s self-concept in cystic fibrosis. J Dev Behav Pediatr. 1981;2(1):5–8.

55.

Marshall

, Sheppard

. Parent participation in paediatric physiotherapy home programmes. Aust J Physiother. 1981;27(6):179–83. doi: 10.1016/S0004-9514(14)60757-1

Validity Content or construct	Rate 2 points if Validity is reported in the paper as tested and the r value or equivalent is≥0.7 or≥ + 7/10 on COSMIN^$.	Rate 1 point if Validity is reported in the paper as tested and the r value or equivalent is≥0.55 but < 0.7 or ≥+ 5/10 but < + 7/10 on COSMIN^$.	Rate 0 points if Validity is reported in the paper as tested and the r value or equivalent is < 0.55 or < 5/10 on COSMIN^$or the paper did not report validity testing and there is no reference to another study reporting validity testing.
Reliability Test-retest reliability or inter-rater reliability or internal consistency reliability	Rate 2 points if Reliability is reported in the paper as tested and the r value or equivalent is≥0.7^*.	Rate 1 point if Reliability is reported in the paper as tested and the r value or equivalent is≥0.55 but < 0.7^*.	Rate 0 points if Reliability is reported in the paper as tested and the r value or equivalent is < 0.5 or the paper did not report reliability testing and there is no reference to another study reporting reliability testing.
Interpretability Responsiveness to change	Rate 1 point if Interpretability is clearly and meaningfully described in the study or there is reference to another publication in which interpretability is clearly described or interpretability is inherently clear from the measure of adherence, e.g., units of time or a scale, etc.	Rate 0 points if The paper did not clearly and meaningfully describe interpretability and there is no reference to another study in which interpretability is clearly described and Interpretability is not inherently clear from the measure of adherence.
Dyad Adherence measure incorporates dyadic considerations	Rate 1 point if The adherence measure considers both the caregiver and the child.	Rate 0 points if The adherence measure considers either the caregiver or the child but not both.