This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on http://www.researchprotocols.org, as well as this copyright and license information must be included.
Families living with chronic or long-term conditions such as chronic kidney disease (CKD), stages 3-5, face multiple challenges and respond to these challenges in various ways. Some families adapt well while others struggle, and family response to a condition is closely related to outcome. With families and professionals, we developed a novel condition-specific interactive health communication app to improve parents’ management ability—the online parent information and support (OPIS) program. OPIS consists of a comprehensive mix of clinical caregiving and psychosocial information and support.
The purpose of this study was to (1) assess feasibility of a future full-scale randomized controlled trial (RCT) of OPIS in terms of recruitment and retention, data collection procedures, and psychometric performance of the study measures in the target population, and (2) investigate trends in change in outcome measures in a small-scale RCT in parents of children with CKD stages 3-5.
Parents were recruited from a pediatric nephrology clinic and randomly assigned to one of two treatment groups: usual support for home-based clinical caregiving (control) or usual support plus password-protected access to OPIS for 20 weeks (intervention). Both groups completed study measures at study entry and exit. We assessed feasibility descriptively in terms of recruitment and retention rates overall; assessed recruitment, retention, and uptake of the intervention between groups; and compared family condition management, empowerment to deliver care, and fathers’ involvement between groups.
We recruited 55 parents of 39 children (42% of eligible families). Of those, about three-quarters of intervention group parents (19/26, 73%) and control group parents (22/29, 76%) were retained through completion of 20-week data collection. The overall retention rate was 41/55 (75%). The 41 parents completing the trial were asked to respond to the same 10 questionnaire scales at both baseline and 20 weeks later; 10 scores were missing at baseline and nine were missing at 20 weeks. Site user statistics provided evidence that all intervention group parents accessed OPIS. Analysis found that intervention group parents showed a greater improvement in perceived competence to manage their child’s condition compared to control group parents: adjusted mean Family Management Measure (FaMM) Condition Management Ability Scale intervention group 44.5 versus control group 41.9, difference 2.6, 95% CI -1.6 to 6.7. Differences between the groups in the FaMM Family Life Difficulty Scale (39.9 vs 36.3, difference 3.7, 95% CI -4.9 to 12.2) appeared to agree with a qualitative observation that OPIS helped parents achieve understanding and maintain awareness of the impact of their child’s condition.
A full-scale RCT of the effectiveness of OPIS is feasible. OPIS has the potential to beneficially affect self-reported outcomes, including parents’ perceived competence to manage home-based clinical care for children with CKD stage 3-5. Our design and methodology can be transferred to the management of other childhood conditions.
International Standard Randomized Controlled Trial Number (ISRCTN): 84283190; http://www.controlled-trials.com/ISRCTN84283190 (Archived by WebCite at http://www.webcitation.org/6TuPdrXTF).
Children and young people (children) aged 0-19 with conditions such as chronic kidney disease (CKD), stages 3-5, often require treatments at home, which can be complex and intrusive. Research into long-term or chronic childhood (hereafter referred to as chronic) conditions helps us understand how families manage the child’s condition at home with remote support from multidisciplinary teams (MDTs) [
Chronic kidney disease stages 3-5 (adapted from the Renal Association).
CKD stage (GFRa) | Clinical criteria | Clinical management |
3a (35-59%) | Moderately reduced kidney function | Observation, control of blood pressure, and risk factors for progression to stage 4 |
3b (30-34%) | ||
4 (15-29%) | Severely reduced kidney function | Planning for end stage renal failure |
5 (<15% or on dialysis) | Very severely reduced kidney function or end-stage renal failure | Treatment choices (renal replacement therapies) |
aGFR=estimated Glomular Filtration Rate.
Families respond in various ways to chronic conditions. Some adapt well to clinical management responsibilities and are able to develop a sense of control over their lives while others struggle to do so. Family response to chronic conditions is closely related to children’s clinical outcomes, and non-adherence to prescribed treatments is the primary cause of treatment failure in conditions such as CKD [
A sense of control has also been associated with the notion of empowerment in pediatric care [
It is important that children with chronic conditions are cared for in ways that minimize emotional trauma and assist in their recovery, and that such ways of delivering care are investigated to see if they are effective [
In addition, family responses to chronic condition management can be affected by individuals’ health literacy skills (ie, the ability to comprehend health information) [
A Cochrane Review [
The current study forms part of a phased-approach to development and evaluation of a complex intervention [
We developed an IHCA, the OPIS application, in collaboration with families and health professionals. The OPIS comprises clinical care-giving support (information on treatment regimens, video-learning tools of MDT professionals explaining how to undertake clinical procedures at home
We implemented and assessed OPIS for feasibility in the kidney unit of a large children’s hospital in the north of England [
The purpose of this paper is to build on our previous report [
The results reported here will inform the development and implementation of a future RCT that will be sufficiently powered to detect significant change in outcomes attributable to OPIS. To our knowledge, OPIS is the first IHCA to have been rigorously developed with families living with CKD and health professionals, and then tested with parents.
Online Parent Empowerment Model in CKD management [
To achieve the stated objectives, we undertook a small-scale study that used a two-group RCT design with data collected at two points in time: entry to the study at baseline and 20 weeks later (ISRCTN: 84283190). Approval to conduct the study was obtained from the National Health Service (NHS) Research Ethics Committee (REC) (Reference: 11/N/W/0268) and the NHS Trust Research and Development department. No incentives were offered to parents for enrolling in the study.
Parents who provided written informed consent were randomly assigned to one of two treatment conditions: (1) usual support involving discussions with members of the MDT when the child was an in- or out-patient, and for children with stage 5 CKD, home visits from a specialist nurse to teach or reinforce clinical skills, as required (control group), or (2) usual support plus password-protected access to OPIS for 20 weeks, which allowed sufficient time for participants to become familiar with OPIS (intervention group). Children at stage 5 CKD require peritoneal dialysis treatment; the specialist nurse teaches the child and family how to perform this treatment over a series of home visits until they are considered competent to perform it on their own. Future home visits follow to ensure the treatment is being performed correctly. Other clinical skills including giving injections, and managing nasogastric or gastrostomy feeds would also be taught at home with ongoing support to ensure competency.
OPIS was housed on a university Web server and accessible to intervention group parents via their personal computers, mobile phones, tablets, or smartphones. A screenshot of one view of OPIS is shown in
OPIS is Health on the Net (HON) certified (HONConduct443339) [
Screenshot of OPIS homepage.
Parents were considered eligible for this study if their child aged 0-19 years was receiving care at the study site, they had not participated in the development of OPIS and they had access to the Internet via a personal computer or mobile device. Eligible parents were notified of the study by a member of the MDT. Interested parents were referred to the researcher appointed to manage the project and collect/analyze data who then explained study requirements, answered any questions, and obtained written adult consent. As fathers and mothers may have differing information and support needs when their child has a chronic condition [
Data were collected between September 2012 and September 2013. Measurements were conducted before parents were randomized. Those in the intervention group received a username and password to enable OPIS access, and after 20 weeks at which point their access to OPIS ceased. Data collection took place at a time/place convenient to the parents, either in the family home or a quiet area in the hospital; one interview was conducted by telephone at the parent’s request [
At baseline, we used an investigator-devised form to collect background data (child age, sex, postal code, CKD stage at study entry/exit; parent age, sex, race/ethnicity, language, educational achievement, socioeconomic status of neighborhood based on postcode, ethnicity, and clinical care experience). At both baseline and 20 weeks, we administered a set of standardized measures in the following order: the Rapid Estimate for Adult Literacy in Medicine (REALM) [
Parent health literacy was measured using the REALM. The purpose of this was to determine if parents were likely to need help with self-administration of the outcome measures. This assessment requires the parent to read aloud a list of 66 generic clinical words (such as “fat” or “impetigo’) arranged in increasing order of difficulty. The score is calculated by awarding one point for each correctly pronounced word and nil for each mispronounced or skipped word. A score of 59 or less indicates low health literacy while a score of 60 or more indicates adequate health literacy. The REALM has face, criterion, and construct validity for use as a health literacy screening tool in the United Kingdom [
Parent management ability was measured using the FaMM. The FaMM was developed to measure how families manage caring for a chronic condition and the extent to which they incorporate management into family life. The FaMM has 53 items overall, with 45 items for all parents and eight additional items for partnered parents only. Items are scored from 1-5, meaning strongly disagree to strongly agree. There are five summated scales for all parents measuring the dimensions of Child’s Daily Life, Condition Management Ability, Condition Management Effort, Family Life Difficulty, and View of Condition Impact as well as a sixth scale only for partnered parents measuring the dimension of Parental Mutuality.
The FaMM Condition Management Ability Scale (12 items) addresses parents’ perceptions of their competence at taking care of the child’s condition. Because the intent of the OPIS intervention was to enhance parents’ ability to manage their child’s CKD, we were especially interested in the Condition Management Ability scale of the FaMM. Higher values mean parents view themselves as more capable of managing the condition. Example items that help to illustrate the concept and its domains include (1) “We have some definite ideas about how to help our child live with the condition”, and (2) “We have not been able to develop a routine for taking care of our child’s condition” [
Parent empowerment was measured using the Service System Subscale of the FES that explores parents’ relationships with health professionals and parents’ level of comfort in asking questions and voicing their opinions [
Father support for managing the child’s CKD was measured by the DADS, a 24-item Likert-type scale with separate forms for mothers and fathers. The DADS was developed to assess the support offered by fathers, and mothers’ perceptions of the quality of that support. The results of confirmatory factor analysis provide support for the construct validity of the DADS, and two factors (amount and helpfulness of fathers’ involvement) best accounted for participants’ responses [
Data were analyzed using IBM SPSS Statistics Version 20. Participants who dropped out were not contacted further in keeping with REC approval, and the data they provided were compared with those of participants who completed the study. Scores on the outcome measures were calculated and missing values on items handled according to the methods prescribed by the developers. Consistent with the nature of the study and small sample size, our post-intervention analyses should be interpreted conservatively. Intraclass correlation coefficients (ICC) were estimated for each outcome measure to assess the level of within-family variation, and model performance was measured by estimating the square of Pearson’s correlation between actual and predicted values. Confidence intervals were estimated for recruitment and retention rates. The internal consistency of all outcome measures was estimated using Cronbach alpha at baseline across the two groups combined for those completing the study, as the outcomes were measured on a population that had not been previously assessed.
Analysis of trends in change in outcome measures from 20 weeks was adjusted for baseline scores using linear mixed models to allow for having more than one parent participating in a family with adjustments for stratification by CKD stage (3 vs 4/5) and ethnicity (white/black vs South Asian). Confidence intervals and effect sizes for the adjusted differences in means at 20 weeks were estimated—the aim in this feasibility study being to inform sample size estimates for a future full-scale RCT rather than to detect significant differences [
A total of 94 eligible children were identified at baseline. A CONSORT diagram (
At baseline, the two groups were balanced in terms of the stratification variable ethnicity (white/black/Afro-Caribbean vs South Asian) but not CKD Stage (3 vs 4/5). No parent who recorded a REALM score of less than 60 (ie, low level of health literacy) at baseline withdrew from the study before trial end.
A quarter of participants (14/55, 25%) were not retained through trial end, including seven from each treatment group. At enrollment, participants were told they could withdraw from the study at any time without providing a reason. However, most parents who withdrew apologized and volunteered a reason (see
CONSORT diagram showing participant flow through study.
Parent and child characteristics by randomization group at baseline for those completing trial.
Characteristic | Control group | Intervention group | |
Participants, n | 22 | 19 | |
Index children, n | 16 | 14 | |
Parent age, mean (SD) | 44.1 (8.3) | 42.7 (10.3) | |
|
|||
|
Female | 11 (50) | 11 (58) |
|
Male | 11 (50) | 8 (42) |
|
|||
|
White European | 16 (73) | 16 (84) |
|
Afro-Caribbean | 0 (0) | 1 (5) |
|
South Asian | 6 (27) | 2 (11) |
|
|||
|
English | 19 (86) | 19 (100) |
|
Bengali | 2 (9) | 0 (0) |
|
Polish | 1 (5) | 0 (0) |
Parent socioeconomic statusa, median (range) | 13,041.5 (96-28,036) | 21,547 (235-29,472) | |
Child age in years, mean (SD) | 10.2 (5.7) | 9.1 (5.5) | |
|
|||
|
Female | 5 (31) | 4 (29) |
|
Male | 11 (69) | 10 (71) |
|
|||
|
Stage 3 | 3 (19) | 6 (43) |
|
Stage 4 | 5 (31) | 0 (0) |
|
Stage 5 | 8 (50) | 8 (57) |
aPostal code-based neighborhood ranking (1=highest level of deprivation to 32,482=lowest deprivation).
In total, 19 parents accessed OPIS with a mean of 23.3 visits per user (SD 20.8, range 2-64); OPIS was visited 443 times with a total of 3154 page visits. The mean duration of time spent on the site per visit was 12 minutes, 11 seconds (range 2 seconds to 58.0 minutes). Visits lasted between 10 and 30 minutes, and 88.9% (394/443) of visitors used desktop/laptop computers, 7.9% (35/443) used mobile phones, and 3.2% (14/443) used tablets. Tablet users spent the longest time on OPIS while the desktop/laptop users spent the shortest. The most common depth of visit (25.5%) entailed viewing 20 or more different screens, with parental viewing in 34.6% of these visits ranging from 9-19 different pages. The highest number of page views in 1 week was 541. There were two peaks in usage: one at the start of the study, and the other when users were notified that new research reports regarding CKD were placed in OPIS by the project team. The most popular area was “Kidney Health” followed by “Case Studies” and “What to eat/drink”. The least popular area was “FAQ”. Average time for page download was 1.8 seconds [
During parent exit interviews, the issue of usage feasibility was frequently highlighted by parents [
In addition, parents found OPIS easy to access and valued the section that describes the roles and responsibilities of the different MDT members responsible for their child’s overall CKD management: “Because it’s always nice that if you go into somewhere knowing you’ve got a picture of a face with a name, you think ah, yeah, we know her” [parent/056/father].
Parents found the range of information on OPIS interesting: “I enjoyed having a good look around it, I found it interesting for somebody that’s involved in it [CKD management] as a parent” [parent/052/mother].
Furthermore, parents appreciated the links provided to other related websites that had been validated by the OPIS research team, including MDT professionals: “Really liked trustworthy links page, knowing the MDT has agreed to them, it really helps, knowing it’s the right information and not scaring you half to death!” [parent/045/mother].
The REALM required self-administration for assessment of health literacy. Parents were given the option to self-administer the other study measures or have the researcher read the questions aloud and record the parent’s response. At baseline, 96% (53/55) of the parents opted to self-administer the other measures, and for expediency some chose to do this while waiting for their child’s outpatient appointment. Parents offered two main reasons for preferring self-administration: either (1) they wanted to complete data collection as quickly as possible because of time constraints imposed by their child’s clinical demands and their own personal commitments, or (2) reading the questions themselves helped them to better understand the issues being explored and to consider their response.
Observing while parents self-administered the REALM enabled the researcher to discreetly determine whether a respondent may have difficulty completing the remaining measures without assistance. Most parents had little difficulty completing REALM, but two fathers struggled to read out several words. In these instances, the researcher adopted an encouraging and reassuring tone, explaining that our purpose was not to judge parents but to help us learn from parents how best to explain clinical terms.
Although flexibility and respect for parents’ preferences regarding place and time for data collection is important, the outpatient waiting area was not always appropriate for this purpose. For example, with the two fathers who displayed discomfort with REALM, the researcher suggested a move to a more private location nearby. However, even after the change of location these fathers both read out the words in an increasingly lower register, with a bowed head and constricted body language. In one instance, the researcher stopped REALM administration before it was completed because of the father’s profound reading difficulty and apparent discomfort. The father went on to complete the remaining measures with the researcher reading questions aloud to him.
At baseline, members of the control group had the four lowest REALM scores (10, 42, 43, and 43); a member of the OPIS group had the fifth lowest score (57). The REALM score at both baseline and 20 weeks was recorded for 82% (18/22) of parents in the control group and 95% (18/19) of parents in the intervention group. Of these parents, 86% (31/36) scored 60 or above (denoting an adequate level of health literacy) at both time points. The remaining 5 parents scored less than 60 (denoting inadequate health literacy) at both time points and 4 of these parents were in the control group.
Administering the FaMM (the measure with the next highest number of items) immediately after the REALM meant that the researcher could reassure parents that the remaining measures (FES and DADS) would not be as time consuming to complete. After completing the FaMM, all parents volunteered that it had helped them to appreciate the amount and level of clinical care they provided for their child. They also said the FaMM prompted them to reflect on issues such as their child’s education and well-being, and how they enabled their child to achieve a good quality of life. While completing the FaMM, two mothers became emotionally distressed as they recalled the burden of care management; the researcher offered to suspend the interview if it was becoming burdensome or to arrange a meeting with a member of the MDT for counseling. Both mothers declined this offer citing the therapeutic benefit for them of processing these emotions and that through participating they were making a positive contribution to future management by other families. A few parents (8/55, 15%) parents stated that some items on the FaMM (such as “It seems as if our child’s condition controls our family life” and “Our child’s condition requires frequent hospital stays”) were not appropriate to their current situation.
The Service System subscale of the FES appeared to be easily understood by parents as no clarification was requested. The DADS scale was also easily understood.
The majority of the outcomes measures had an acceptable level of internal consistency reliability; a Cronbach alpha ≥.70 is commonly considered as being acceptable in psychosocial research. The two exceptions were the FaMM Condition Management Ability Scale (alpha=.52) and Condition Management Effort Scale (alpha=.62). In the Condition Management Ability Scale, the item “We have enough money to manage our child’s condition” was negatively correlated with eight of the other 11 items, and its correlation with the sum of the other items was -.21. Deleting this item from the scale increased Cronbach alpha from .52 to .62. Another two items, “We have some definite ideas about how to help our child live with the condition” and “We often feel unsure what to do regarding our child’s condition”, showed very little correlation (
Parents had not needed several items in the DADS during the previous 6 months. For example, 20 parents had not attended a support group or educational workshop about their child’s condition and had not needed to pay medical bills. Only two parents had complete entries for the Amount score.
Descriptive statistics and reliability estimates for the study outcome measures at baseline across total sample for parents completing the trial.
Outcome measure | Items, n | Complete responses, n | Number scored, n | Percentage relevant scored, % | Mean (SD, range) | Cronbach alphac | |
|
|||||||
|
Child’s Daily Life Scale | 5 | 39 | 41 | 100.0 | 17.3 (5.0, 10-25) | .72 |
|
Condition Management Ability Scale | 12 | 41 | 41 | 100.0 | 45.0 (5.9, 27-56) | .52 |
|
Condition Management Ability Scale (revised)a | 11 | 41 | 41 | 100.0 | 43.6 (5.4, 32-55) | .62 |
|
Condition Management Effort Scaleb | 4 | 41 | 41 | 100.0 | 14.0 (3.6, 6 -20) | .62 |
|
Family Life Difficulty Scaleb | 14 | 40 | 41 | 100.0 | 36.1 (12.3, 14-56) | .90 |
|
Parental Mutuality Scale | 8 | 35 | 35 | 100.0 | 33.4 (6.2,19-40) | .79 |
|
View of Condition Impactb | 10 | 38 | 41 | 100.0 | 30.1 (6.3,14-41) | .69 |
|
|||||||
|
Service System Subscale | 12 | 38 | 39 | 95.1 | 4.2 (0.5, 3.1-5) | .85 |
|
|||||||
|
Amount score | 24 | 33 | 33 | 94.3 | 79.2 (21.2, 50.1-120.0) | .91 |
|
Helpfulness score | 24 | 32 | 34 | 97.1 | 73.0 (19.9, 33.0-112.0) | .95 |
aExcluding the contradictorily correlated item “We have enough money to manage our child’s condition”.
bHigher scores are undesirable.
cBased on complete responses for each scale.
Outcome scores at baseline by randomized group for parents completing the trial.
Outcome measure | Control group (n=22) | Intervention group (n=19) | |||||
n | Mean (SD) | Range | n | Mean (SD) | Range | ||
|
|||||||
|
Child’s Daily Life Scale | 22 | 17.6 (5.6) | 10-25 | 19 | 17.0 (4.4) | 11-25 |
|
Condition Management Ability Scale (revised) | 22 | 42.8 (4.8) | 36-52 | 19 | 44.5 (6.0) | 32-55 |
|
Condition Management Effort Scalea | 22 | 13.1 (3.9) | 6-20 | 19 | 14.9 (3.0) | 8-20 |
|
Family Life Difficulty Scalea | 22 | 35.3 (13.8) | 14-56 | 19 | 37.0 (10.6) | 15-56 |
|
Parental Mutuality Scale | 19 | 32.8 (6.3) | 19-40 | 16 | 34.1 (6.1) | 20-40 |
|
View of Condition Impacta | 22 | 30.3 (6.0) | 20-41 | 19 | 29.8 (6.8) | 14-40 |
|
|||||||
|
Service System Subscale | 17 | 4.3 (0.5) | 3.1-5.0 | 16 | 4.1 (0.5) | 3.2-4.9 |
|
|||||||
|
Amount score | 17 | 84.4 (22.5) | 50.7-120.0 | 16 | 73.9 (18.7) | 50.1-115.2 |
|
Helpfulness score | 18 | 69.9 (21.1) | 41.8-102.0 | 16 | 76.6 (18.4) | 33.0-112.0 |
aHigher scores are undesirable.
Estimated marginal meansa for outcome scores by randomized group and their differences at end trial using linear mixed model.
Outcome measure | Control group |
Intervention group |
Intervention minus Control | ||||||
n | Meana (95% CI) | n | Meana (95% CI) | Diff (95% CI) |
|
|
ICC | ||
|
|||||||||
|
Child’s Daily Life Scale | 22 | 16.9 (13.8-19.6) | 19 | 15.7 (12.8-18.7) | -0.9 (-4.3 to 2.5) | .576 | .856 | .568 |
|
Condition Management Ability Scale (revised) | 22 | 41.9 (38.5-45.4) | 19 | 44.5 (40.9-48.1) | 2.6 (-1.6 to 6.7) | .213 | .823 | .440 |
|
Condition Management Effort Scaleb | 22 | 13.3 (11.0-15.6) | 19 | 15.2 (12.8-17.6) | 1.8 (-0.9 to 4.6) | .176 | .613 | .247 |
|
Family Life Difficulty Scaleb | 22 | 36.3 (29.0-43.5) | 19 | 39.9 (32.5-47.3) | 3.7 (-4.9 to 12.2) | .389 | .937 | .778 |
|
Parental Mutuality Scale | 19 | 31.0 (27.7-34.3) | 16 | 34.8 (31.4-38.2) | 3.8 (-0.3 to 7.9) | .066 | .421 | .138 |
|
View of Condition Impactb | 22 | 29.9 (26.1-33.7) | 19 | 30.6 (26.8-34.4) | 0.7 (-3.8 to 5.1) | .763 | .953 | .829 |
|
|||||||||
|
Service System Subscale | 21 | 4.3 (4.0-4.6) | 18 | 4.2 (3.9-4.5) | -0.2 (-0.5 to 0.2) | .404 | .803 | .456 |
|
|||||||||
|
Amount score | 17 | 78.1 (61.3-94.8) | 16 | 73.8 (57.5-90.2) | -4.3 (-24.7 to 16.2) | .667 | .794 | .614 |
|
Helpfulness score | 18 | 70.0 (60.9-79.0) | 16 | 82.3 (72.6-91.9) | 12.3 (0.9-23.7) | .036 | .211 | .161 |
aEstimated marginal mean adjusted for baseline score, severity of chronic kidney disease, ethnicity, and number of parents in family.
bHigher scores mean worse outcomes.
After 20 weeks, parents in the intervention group had an adjusted mean score on the FaMM Condition Management Ability Scale (revised) that was 2.7 points better than that for parents in the control group (95% CI -1.6 to 6.7). The linear mixed model for this outcome was a good fit to the patterns in the data (
Differences in marginal means between the study groups in terms of whether the difference favored the intervention (OPIS) group or the control group (means adjusted for baseline score, severity of chronic kidney disease, ethnicity, and number of parents in family).
The main findings of this study are that a full-scale RCT of the effectiveness of OPIS is feasible and that OPIS has the potential to beneficially affect self-reported outcomes, including parents’ perceived competence to provide home-based clinical care for children with CKD stage 3-5. In this section, we first address objective 1 by assessing feasibility of a future full-scale RCT of OPIS. Then we address objective 2 by considering the trends in change on study outcome measures.
The results support the feasibility a full-scale RCT. However, in future research recruitment could be improved by examining the potential influences on recruitment to this study. For example, parents could view participating in a study relating to their child’s health care with no promise of benefit to themselves or their child, as adding burden to an already stressful situation. Due to the complex care needs of many children with CKD, in particular those with level 4/5 CKD, the unpredictable nature of individuals’ disease progression and the potential for the child’s condition to deteriorate during the trial, some parents might have declined to participate because of time requirements and duration of commitment.
The researcher recruiting parents and collecting and analyzing data was not a health professional, so not a member of the MDT. Parents were unlikely to feel obliged to agree to the study when approached by the researcher in the way they might have if recruited by a MDT member. We believe it important that the MDT was seen to endorse the research and the researcher. This happened through our strategy of arranging for a professional to introduce the parents to the study during a clinical consultation. This was important because parents often have a long-standing and trusting relationship with members of the MDT and so may have wished to discuss the study with them before making a decision about participation. In future research, more proactive MDT endorsement of the study, such as by referring to OPIS and/or demonstrating it when providing specific information to parents about the child’s condition, could further enhance recruitment.
The study design required that parents be randomly allocated to a treatment group. Some parents who were initially interested in the study could have declined enrollment once they realized there was no certainty of allocation to OPIS. Parents possibly also felt an obligation to express interest in the study to the professional who notified them of the study but later felt able to tell the researcher that they were not interested once they realized that their refusal would not be reported to the MDT. In a future full-scale trial, we can adopt a number of additional strategies to potentially increase recruitment/retention. For example, parents usually have several individual outpatient consultations with members of the MDT at one appointment. We capitalized on the opportunity presented by this as, once the first professional had notified the parents of the study, the researcher then approached them to explain more about what would be involved. Some parents might have been too distracted by their child’s presence and the other pending consultations to be able to give due consideration to the possibility of participating and may have preferred an explanatory telephone call from the researcher at a later date. To enable this, the MDT member who initially informed the parents about the study would also need to ask whether parents would consider either an explanatory phone call from the researcher at a later date or a face-to-face meeting with the researcher in clinic on the same day.
In addition, we could alter the study design so that randomization would be to either the intervention (OPIS) or to a wait-list control group. The wait-list control group could elect to receive the intervention immediately after the post-test assessment [
At the initiation of baseline data collection, the REALM was a useful tool to help the researcher determine whether a parent appeared to struggle with reading. This meant the researcher could alter the data collection strategy accordingly to minimize embarrassment for parents with low health literacy and maximize the quality of the data. The measurements of health literacy were stable from baseline to end of trial, which is understandable since health literacy is a relatively stable construct and OPIS was not intended to improve health literacy. The REALM proved to be difficult for 2 parents due to its large number (66) of items. A recently validated Rapid Estimate of Adult Literacy-Short Form (REALM-SF) comprising only seven items could be more appropriate for a future full-scale RCT of OPIS [
Some potential reasons for attrition in the intervention group are that (1) we adopted a non-directive strategy in that once intervention group parents had received the password and login advice, and (2) we did not direct parents’ use of OPIS as we wished to determine parents’ undirected usage. Parents might have expected more direct and continued engagement with the study team rather than self-directed exploration of OPIS. In addition, parents might have expected more endorsement by MDT members than was possible within the study resources; indeed our qualitative interviews with parents at study exit confirm this [
Retention in research that involves Internet interventions has been identified as a major problem and is now widely recognized as a science of attrition [
While the numbers dropping out of our study for the different measures were between 4 and 7 parents, patterns of attrition tended to differ between the treatment groups. Although the attrition pattern in the intervention group was inconsistent, those dropping out tended to perceive more problems with family life and more impact caused by their child’s condition at baseline than those intervention group parents who were retained in the study. This finding is consistent with the finding that the scores on these measures were “poorer” in the intervention group than in the control group (see
To address objective 2, we investigated trends in change on outcomes in a small-scale preliminary RCT in parents of children with CKD stages 3-5. The results suggest that OPIS could improve parents’ ability to manage their child’s condition more than standard care over 20 weeks. For example, after accessing OPIS for 20 weeks parents were less likely to endorse the statement “We have not been able to develop a routine for taking care of our child’s condition”. When children have a chronic condition such as CKD, parents usually assume the roles of care coordinator, clinical expert, and advocate as well as their normal parenting roles. Health care providers are uniquely positioned to assist parents in meeting those challenges, and researchers recommend that they aim to promote parents’ competence and confidence in their child's care through understanding common challenges that parents face, promoting parent-to-parent connections, and building partnerships with parents and their children with clinical needs [
The FaMM Parental Mutuality Scale indicated that parents in the intervention group tended to show less decline in satisfaction in working together over time. Reasons for decline in satisfaction should be explored in a larger study. A smaller drop in parental mutuality in intervention group scores on this measure concurred with more desirable change in the DADS Helpfulness scale. A strong correlation between the FaMM Parental Mutuality Scale and the DADS Helpfulness (correlation) subscale suggests that it would not be necessary to include both the FaMM and the DADS in our future full-scale RCT, thus reducing participant burden.
The FaMM View of Condition Impact Scale showed that both groups’ perceived seriousness of the child’s condition tended to show improvement, and more so in the control group as compared to the intervention group.
The FaMM Family Life Difficulty Scale showed that the control group parents appeared to perceive fewer difficulties after 20 weeks, but the intervention group appeared to perceive slightly more. It was not clear whether this was a negative or a positive impact, since OPIS may have elevated parents’ awareness of their caregiving responsibilities. This would need investigating in a larger study using a measure of anxiety as there could be implications for clinical practice.
We also know that adherence to medical recommendations deteriorates over time [
Information management, specifically limiting awareness of information that generates uncertainty, fear, and anxiety has been well established as a typical parental management strategy in the context of life-threatening chronic childhood conditions [
IHCAs such as OPIS might have the potential to increase fathers’ involvement in disease management if suggestions for refinement and usability issues that we reported elsewhere [
A limitation of this study is the low number of participating parents of South Asian descent, despite the relatively high prevalence of CKD in their children. In addition to reasons given by these parents for declining to participate (eg, no time or child transferring to adult care), cultural and language barriers to participation might have been in play. South Asian women, who are often the primary caregivers, can possibly lack knowledge of health risks, have ideas about self-care that differ from those held by white women of European descent, experience language barriers, be subject to the stress of emigration and isolation, be preoccupied with their family’s needs, and not seek access to health promotion programs [
The number of eligible children being cared for at the site during the study period was less than anticipated, and we did not achieve the target of data from 30 parents per group at trial end. The future full-scale RCT could recruit parents from multiple sites, with the intervention being delivered via the Web as in the feasibility study. Overall, the data obtained from our 51 participants were highly informative.
Our results indicate that a full-scale RCT of OPIS is feasible. Furthermore, being in the intervention group improved parents’ perceived management ability to a greater extent than usual care over 20 weeks. Specifically, the FaMM Condition Management Ability Scale appeared to show beneficial change when reinforced by accessing OPIS. A full-scale trial of OPIS is indicated that would include a shorter measurement of health literacy and additional measurements of self-efficacy and anxiety, with an embedded qualitative component to investigate reasons underlying changes in outcome scores.
chronic kidney disease
Dads Active Disease Support scale
Family Management Measure
Family Empowerment Scale
Health on the Net
interactive health communication application
multidisciplinary team
National Health Service
Online Parent Information and Support
Rapid Estimate of Health Literacy
Research Ethics Committee
This paper presents independent research commissioned by the National Institute for Health Research (NIHR) under the Research for Patient Benefit program (PB-PG-0110-21305). The views expressed in this publication are those of the authors and not necessarily those of the NHS, the NIHR, or the Department of Health.
We would like to thank the following for their help: the parents, children, young people, and professionals who generously provided data, support, and advice; members of the steering group who supported the wider study, including Dr Hanif and Mr Osman—both parents of children with CKD—who acted as expert consumer advisors to the research team; Melanie Harper-Jones, Professor Nick Webb, and Professor George Knafl for valuable contributions to the study.
None declared.