Parent-held child health records do not improve care: A randomized controlled trial in Norway

Abstract

Objective. To study the effects of a parent-held child health record (PHCHR) that was created by the Norwegian Board of Health with the purpose of introducing this to the whole country. Design. Randomized controlled trial. Setting. Maternal and child health centres in 10 municipalities in Norway. Subjects. Parents of 309 children attending the National Preschool Health Surveillance Programme. Intervention. Half of the parents were given a PHCHR and short instructions on how it was expected to be used. Main outcome measures. Parent–professional collaboration, healthcare utilization, and parents’ knowledge about child health matters and illness. Results. Some 73% of the intervention group used the PHCHR regularly when visiting the health centres, 79% reported that their own writing in the record was helpful, and 92% favoured the PHCHR being permanently adopted. Use of the record did not influence the utilization of healthcare services, parents’ knowledge of their child's health, or parents’ satisfaction with information or communication with professionals. Conclusions. The PHCHR was well accepted by parents and professionals but it had no effects on collaboration, healthcare utilization, or other measures that could justify the costs of introducing the record into common use. Therefore, the introduction of a parent-held child health record in Norway is being postponed.

• Child health services
• general practice
• health surveillance
• medical records
• parent-held child health record
• parents

Parent-held child health records (PHCHR) have been promoted as an effective tool for giving information to parents and for collaboration between professionals regarding children's health and development [1]. In some countries, parent-held records have been in use for more than two decades [2]. Previous evaluations have concluded that PHCHR are useful for both parents and professionals [3–5].

Norway has a national health surveillance programme that includes immunization for preschool children and enjoys high approval, with participation close to 100%. Well-child visits take place in local maternal and child health centres staffed by public health nurses and part-time general practitioners (GPs) and physiotherapists. Specially designed joint-professional child health records are kept at the centres. There is no systematic sharing of information with other health agencies. Except for a few local, non-published initiatives, there is no common PHCHR in regular use in Norway.

Parent-held child health records have been promoted as an effective tool for giving information to parents and for collaboration between professionals concerning children's health.

• As in other countries, parent-held child health record would be well accepted by parents and professionals in Norway.

• Using a randomized controlled study design, parent-held child health records had no effect on parent–professional communication, healthcare utilization, or parents’ knowledge about child health matters.

Based on this background, Norwegian health authorities established a working group to develop a PHCHR intended to enhance health promotion and parent participation. Prior to its nationwide use, a trial of the parent-held record was conducted to determine to what extent it would:

• be used in child health services as a whole – not only in well-child visits;

• increase parents’ satisfaction related to participation and communication with professionals;

• influence healthcare utilization;

• enhance parents’ knowledge concerning child health matters and illness.

The trial was to be carried out within the standard health surveillance programme with ordinary staff.

Material and methods

A geographical area comprising 10 municipalities in More and Romsdal county with 14 child health centres, both urban and rural, was selected. The study population was all children six weeks to five years old (born between 1 August 2000 and 1 October 2001) who came to one of the centres for a well-child visit. The public health nurse asked the parents for their consent to participate in the trial and their names were sent to the project leader. Firstborn children were excluded because their parents would not have previous experience with the traditional routines to compare with. Enrolled children were randomized into one of two groups by simple drawing (Figure 1). We had previously calculated that 300 subjects were needed to detect a difference of clinical importance that we decided to be at least 15% between the two groups (p < 0.05) [6]. Information about who belonged to the intervention group and the control group then was sent to the child health centres.

Figure 1.  Random allocation of children into equal-sized intervention and control groups.

Baseline data were collected with a self-administered questionnaire given at the beginning of the study and again one year later. Parents were asked about utilization of healthcare services, knowledge of their child's health, and their satisfaction with communication with healthcare professionals. Most questions were collected from previously surveys [7] and we tested the questionnaire in a pilot study. The intervention group was also given the new PHCHR and short instructions on how it was expected to be used. This group's exit questionnaire included some additional questions about the parent-held record. The new PHCHR had been designed by a working group appointed by the health authorities who had studied PHCHRs from other countries. It included appointments, vaccinations, steps in child development, room for both health personnel and parents to make notes, and more discussed below.

Change in the response variable from baseline to a later point in time is considered an adequate outcome measure when evaluating the effect of an intervention trial, mainly since it may reduce potential bias due to baseline imbalance between the intervention and control group [8]. Therefore, we calculated the difference between parents’ baseline and follow-up response to specific questions related to the three main topics mentioned above (i.e. difficulties communicating with healthcare professionals, agreement in knowledge of child's health, and frequency in use of healthcare services). The participants were subsequently categorized into three groups based on these calculated differences, indicating whether the parents had reported fewer, same, or more difficulties/agreement/frequency between the first and the second measurement. We used Pearson's chi-squared test to evaluate whether the intervention group and the control group changed differently between baseline and follow-up and, hence, if the intervention (i.e. having a PHCHR) may have had any effect. However, due to small expected numbers in one of the analyses, we used Fisher's exact test to calculate the p-value. All analysis was conducted using SPSS for Windows, version 12.

The study design was approved by the Norwegian Regional Ethical Committee in Trondheim.

Results

All but one parent agreed to participate, and so 309 children were enrolled. In total, 94% of parents returned the first questionnaire and 84% returned the second, one year later; 97% of questionnaires were filled out by mothers. There were no differences of significance between the intervention and control groups in child sex, age, presence of chronic illness, parental education, or parental marital status. The age distribution is given in Table I.

Table I.  Number and percentage (%) of children in each age group at the time of intervention.

	Children's age in months
	0–2	3–5	6–8	9–11	12–14	>15	Total
Intervention	20 (13)	24 (16)	48 (32)	30 (20)	27 (18)	1 (1)	150
Control	12 (9)	42 (30)	34 (25)	23 (17)	25 (18)	4 (2)	139

Use of the parent-held child health record in different healthcare services

Parents were encouraged to bring the PHCHR to all healthcare encounters. Figure 2 shows that PHCHRs were used only to a limited extent in encounters outside the child health centres. Few took the record with them for specialist healthcare services. About 40% brought the record either regularly or occasionally when visiting their GP; 89% of parents said that the record would have been used more if professionals had shown more interest. We found a positive correlation between the use of the record and both frequent healthcare encounters and having a child with a chronic illness.

Figure 2.  Percentage of parents bringing the parent-held child health record with them when visiting different healthcare services.

Parent participation and communication with professionals

Only a minority of parents in the intervention group said that the parent-held record had enhanced healthcare participation (31%) or communication (39%) compared with their earlier experiences. In all, 53% said that the record had given them a better overview of their child's health. Table II shows that the parent-held record did not make a difference in the difficulty parents felt when talking to professionals. Close to 80% of all parents reported no difficulties in asking professionals questions about their child. Parental satisfaction with information provided about their child's health from different professionals was also the same in both groups.

Table II.  Change in experience during a period of one year regarding difficulty in talking to health personnel.

	Control	Intervention
Variables	N (%)	N (%)	p-value^a
Nurse at the centre
Less difficult	18 (14)	14 (12)
Same	86 (77)	97 (82)
More difficult	11 (9)	8 (7)	0.66
Doctor at the centre
Less difficult	13 (11)	10 (9)
Same	92 (75)	89 (75)
More difficult	17 (14)	19 (16)	0.78
Other doctors
Less difficult	10 (10)	10 (11)
Same	82 (79)	63 (71)
More difficult	12 (12)	16 (18)	0.39
Other health personnel
Less difficult	5 (11)	3 (13)
Same	36 (77)	20 (83)
More difficult	6 (13)	1 (4)	0.60^b

^ap-value from Pearson's chi-squared test. ^bEstimated with Fisher's exact test due to small expected numbers.

Effect on healthcare utilization

While there was a noticeable decline in encounters at the child health centres during the study, Table III shows that this did not differ between the intervention group and the control group. Looking only at children with chronic diseases, 17% more parents in the control than the intervention group visited the child health centre but there was no difference in utilization of specialist care.

Table III.  Change in frequency of contact with healthcare services during a one-year period.

	Control	Intervention
Variables	N (%)	N (%)	p-value^a
Non-routine CHC during three months prior to questionnaire
Less often	11 (9)	7 (6)
Same	76 (62)	82 (66)
More often	35 (29)	35 (29)	.58
Doctor outside CHC during three months prior to questionnaire
Less often	35 (29)	37 (30)
Same	59 (48)	58 (46)
More often	28 (23)	30 (24)	.95
Specialist or hospital during six months prior to questionnaire
Less often	18 (14)	19 (15)
Same	91 (73)	93 (74)
More often	16 (13)	13 (10)	.84

^ap-value from Pearson's chi-squared test. CHC = child health centre.

Parents’ knowledge about child health matters

The questionnaire included 14 questions about mother–child interaction, illness, and child development, answers to which could be found in the parent-held record. Table IV gives some examples. Overall, parents’ knowledge was quite good and increased during the study, but there was little difference between the two groups.

Table IV.  Change in parents’ knowledge concerning their child's health during a one-year period.

	Control	Intervention
Variables	N (%)	N (%)	p-value^a
Newborns get the best view of me at a distance of 60 cm
Agree less	37 (32)	34 (30)
Same	55 (48)	59 (52)
Agree more	23 (20)	21 (18)	.84
I should call a doctor when my child's temperature passes 39°C
Agree less	22 (18)	30 (25)
Same	58 (47)	44 (37)
Agree more	44 (36)	44 (37)	.22
I should let the child use a baby-walker or jumping-reins
Agree less	8 (8)	8 (7)
Same	95 (79)	99 (86)
Agree more	17 (14)	10 (9)	.40

^ap-value from Pearson's chi-squared test.

Perceived utility of the parent-held record

Some 65% of parents were satisfied with having the parent-held record available, and 92% were in favour of making its availability permanent. Satisfaction and support were especially high among parents of children with chronic diseases. Altogether, 73% of parents had entered additional information in the parent-held record. About half of the parents said they had studied the record specifically at home; 41% said that the record made them more aware of their child's health and development, and 38% said it made it easier to tell healthcare providers about their worries.

Discussion

This study shows that the PHCHR was popular with and perceived as useful by most parents. However, it did not demonstrate any effect on the expected measures: it did not make parents participate or communicate more with professionals, and it did not affect healthcare utilization or parents’ knowledge about child health and development.

A weakness of this study is that, in spite of randomization into two groups, we could not prevent participants in the two groups from exchanging information with each other. Many of the participants lived in the same neighbourhoods and knew each other prior to the study. Also, they met at the child health centres. Information exchanges could mask the observed effect of the parent-held record. Alternatively, we could have randomized the child health centres instead of the children, treating all participants at each centre the same. However, this would have required a far greater number of participants, and could introduce new problems because of clustering. Better training of health professionals in motivating and educating parents on the use of the record might also have also improved its effect. One also has to take into account that routines at the child health centres have become well established over many years and some of the variables measured had a high baseline and as such would be difficult to influence.

The study was limited to a particular geographical area and the results should only be generalized with great caution. However, the organization of and activities at maternal and child health centres are standardized and very much alike all over Norway.

Our results differ from early studies that supported the PHCHR. However, these studies did not have control groups [1–4]. More recent prospective randomized controlled studies are in better agreement with ours: personally held records are strongly approved of by some health professionals and patients, but no effects have been shown on provision of information, patient satisfaction, or collaboration between professionals [9–11]. Our results indicate that children with chronic disease have a different pattern of healthcare utilization but the number was too small to analyse the use of the PHCHR for this group in particular.

We have not found any study that explains the discrepancy between the parents’ and professionals’ perceived effects of personal records and the non-effect found under controlled circumstances. In both the intervention and control groups there are some positive changes in the expected measures over time, which agree with parents’ and professionals’ perceptions, but these are not shown to be caused by the parent-held record. We suspect most of this improvement is due to parents’ growing experience, both of having children and of collaborating with health professionals.

Conclusion

A PHCHR as a supplement to the child health record at maternal and child health centres did not yield the expected advantages. Based on these findings, the Norwegian health authorities have postponed the introduction of a national PHCHR. Further research is needed on parents’ information needs and methods for promoting parent–professional collaboration. Shared electronic child health records, to which both professionals and parents have full access, are one alternative that may be able to improve information exchange and parental participation [12]. One should also take a closer look at PHCHRs specially designed for children with chronic disease.

The authors would like to thank the staff at the participating maternal and child health centres who contributed their time and willingness to make this study possible.