Article Text

Original article
Association between prescribing hypnotics for parents and children in Norway
  1. Ingvild Holdø1,
  2. Marte Handal2,
  3. Svetlana Skurtveit1,2,
  4. Jørgen G Bramness1,2
  1. 1Norwegian Centre for Addiction Research, University of Oslo, Oslo, Norway
  2. 2Department of Pharmacoepidemiology, Norwegian Institute of Public Health, Oslo, Norway
  1. Correspondence to Ingvild Holdø, Norwegian Center for Addiction Research, University of Oslo, PB 1039, 0315 Oslo, Norway;ingvild.holdo{at}


Objective To describe the dispensing of the hypnotic alimemazine to children aged 0–3 years and investigate the association between dispensing of alimemazine to children and dispensed hypnotics to their parents.

Design An observational cohort study linking information from the Medical Birth Registry of Norway and the Norwegian Prescription Database. Hypnotics dispensed to parents in a 1-year period before pregnancy was associated with dispensed alimemazine for children aged 0–3 years.

Patients and setting All children born in Norway in 2008 (N=59 325) and their mothers and fathers were included.

Main outcome measures Dispensed alimemazine to children during the first 3 years of life.

Results Three percent of children received alimemazine. Dispensed hypnotics to mothers increased the risk of the child receiving a prescription for alimemazine, OR of 2.3 (1.7–3.0) for boys and 1.7 (1.2–2.4) for girls. When both parents had been dispensed prescriptions for hypnotics, the risk increased nearly threefold. A dispensed alimemazine prescription was also associated with dispensed prescriptions for antidepressants to both mother and father, mother's smoking, the child's gender and child's prescriptions for antibiotics, respiratory drugs and dermatological steroids.

Conclusions Dispensed alimemazine to children under 3 was associated with parents’ previous use of hypnotics, indicating that factors other than the child's health influence the use of hypnotic drugs in infancy and toddler years. The frequent usage of alimemazine in children below 3 years and the association with parents’ use of hypnotics should concern prescribing doctors.

  • Epidemiology
  • Sleep
  • Pharmacology
  • Paediatric Practice

Statistics from

What is already known on this topic

  • There are concerns regarding both the safety and efficacy of hypnotic drugs such as alimemazine in young children. Despite these concerns alimemazine is widely prescribed for young children in Norway.

What this study adds

  • Three per cent of children less than 3 years old received alimemazine. These children were more likely to have parents who had previously received hypnotic drugs.


The sedative antihistamine, alimemazine, is commonly prescribed to 0–3 year olds.1 For many years, it has been recommended for children with sleep problems,2 and it is almost the only hypnotic drug dispensed to infants and toddlers in Norway.3 This prescribing of alimemazine as a hypnotic to children has been questioned.1 ,4 ,5

Alimemazine is a first-generation phenothiazine antihistamine developed in the 1950s. It easily crosses the blood–brain barrier and probably exerts its sedative action by blocking histamine's neurotransmitter function. Like other phenothiazines, it is a non-specific drug with antagonist action on a variety of receptors (histamine, 5HTserotonin, dopamine, α-adreno and muscarinic receptors) and a multitude of possible side effects.6

Case reports of significant toxicity have been described in children following the use of alimemazine.7 ,8 Phenothiazines have been associated with sudden infant death.9–12 Alimemazine is not recommended for children below the age of 2 years.13

The effectiveness of alimemazine for helping children with sleep problems has been studied in four small randomised controlled trials.14–17 The trials found conflicting results with some reporting short-term improvement only.14–17 Behavioural intervention is often recommended as first line of treatment.18

The relationship between parent and offspring drug use has recently been targeted, and studies have shown an association between mothers and their teenage offspring in their use of opioids and benzodiazepines/z-hypnotics.19 ,20

We wanted to examine the level of dispensed prescriptions of alimemazine to children aged 0–3 years and any association between this and dispensed hypnotic drugs to parents in a longitudinal design. We hypothesised that there would be an association between previously dispensed hypnotic drugs to parents and alimemazine to children below the age of 3.

Material and methods

Data sources

In this cohort study, we linked information from the Norwegian Prescription Database (NorPD) and the Medical Birth Registry of Norway (MBR). The unique person identifier given to all Norwegian inhabitants makes it possible to link different registries at an individual level.

The Norwegian Prescription Database

All pharmacies have, since 2004, been obliged to register dispensed prescriptions to individuals outside institutions electronically and send the information to NorPD.21 Each prescription is registered with detailed drug information, including Anatomical Therapeutic Chemical (ATC) code,22 and patient identifiers.

NorPD provided information on all dispensed alimemazine (ATC-code R06AD01) to children and all dispensed hypnotics (N05C) to the parents. We also received data on antidepressants (N06A) used by the parents and data on other frequently prescribed drugs for children: antimicrobials (J01), dermatological steroids (D07) and drugs for obstructive airway diseases (R03) (see online Web-only Table).

The Medical Birth Registry

Since 1967, the MBR has registered all pregnancies in order to supervise practise and reveal health problems related to pregnancy and birth.23 Information is obtained partially by interview and partially by clinical report from doctors/midwives during pregnancy and at delivery. The registry holds unique identifiers of mother, child and father, and can therefore, be used to couple families.

From the MBR, we received information on all children born in 2008 and their mothers and fathers. We were also given information on the gestational age, the child's gender, birth month, birth year, county of residence, maternal smoking in pregnancy, parents’ parity, mother's marital status and the mother's age at birth. The smoking variable was originally five different variables. These were collected into one variable with the following possible outcomes: smoking yes, no, consent to give information on smoking not given or data missing.

Study population

The file MBR contained 61 900 children born in 2008 and their parents. We excluded cases where the ID was missing. Some mothers had several children in 2008. These children were either born the same month (twins or triplets or quadruplets) or in two different months (two pregnancies). We included all twins, triplets and quadruplets but only the firstborn of the siblings born the same year. Our final study population was 59 325 children, 95.8% of the 2008 birth cohort.

We only had access to birth month and year and created an estimated birth date by setting the birth day to the 15th of the month for all children. The exposure period was a 1-year period before the beginning of pregnancy, found by subtracting the gestational age from the estimated birth date. The outcome period was the first 35 months after the birth date.

We gathered prescription information on parents’ dispensed hypnotics during the 12 month period prior to pregnancy, and alimemazine dispensed to the children from birth until 3 years of age. Information on parents’ and children's dispensed prescriptions for the other medications were also gathered in the exposure and outcome periods, respectively. Individuals were categorised as users of a drug if they had been dispensed at least one prescription in the period.


χ2 test was used to compare differences in relation to gender and users and non-users of alimemazine with a significance level of p<0.05.

The confounders that showed a significant relationship to the outcome were included in a binary logistic regression analysis reporting ORs with 95% CI (95% CI). We allowed for clustering on same mother, thus compensating for mothers having twins, triplets or quadruplets within the observation period. The regression was first done for the variables separately (unadjusted OR) and then the significant variables (mothers’ use of antidepressants, mothers’ smoking for both genders and mothers’ parity for boys) were included in the adjusted analysis. Boys and girls were analysed separately and the outcome was any use or no use of alimemazine. The analysis was performed in a stratified manner with use or no use of antibiotics, respiratory drugs and dermatological steroids (data not shown). When looking at the impact of benzodiazepine and z-hypnotics separately, we also adjusted for the others parent's use of any hypnotic.

All analyses were performed using IBM Statistical Package for the Social Sciences (SPSS) V.20 and STATA V.12.1.


Alimemazine was dispensed to 1744 (3.0%) of the children before the age of 3, 1009 (3.4%) of the boys and 735 (2.6%) of the girls. Most children were dispensed the first prescription during their second year of life (table 1). The vast majority of children were only dispensed alimemazine once, but the proportion of boys with more than two prescriptions was greater than that of girls. Children who used alimemazine were more likely to have been given antibiotics, respiratory drugs and dermatological steroids (table 2).

Table 1

Age and number of prescriptions per child for Norwegian boys and girls born in 2008 receiving alimemazine before 3 years of age

Table 2

Associations between exposures (hypnotic used by parents), all confounders and effect modifiers and prescriptions for alimemazine to Norwegian children 0–3 years of age born in 2008

When mothers or fathers had used hypnotics prior to pregnancy, the proportion of children receiving alimemazine was almost doubled compared to those whose parents had not used hypnotics (table 2). Mothers who smoked more often had alimemazine prescribed to their children. Fifteen percent of the women did not volunteer data on smoking habits and we had missing data for an additional three percent. A primapara mother was less likely than a multipara mother to have a child receiving alimemazine. Mothers with a record of antidepressant use were twice as likely as others to have had alimemazine dispensed to their children.

Results from the logistic regression (table 3) show that parents who had used hypnotics were more likely to have children who used alimemazine. The mother's use was a stronger predictor than the father's use. The strongest association was when both parents had received hypnotics, adjusted OR of 2.3 (1.7–3.0) for boys and 1.7 (1.2–2.4) for girls. Stratified analysis by antibiotic use still showed an association between parents previous drug use and alimemazine to children. When both parents had used hypnotics, the OR of alimemazine use was much larger for the non-antibiotic users. Stratified analysis according to use of dermatological steroids and respiratory drugs showed similar results (data not shown).

Table 3

The OR with 95% CI (95% CI) for boys and girls 0–3 years of age to be prescribed alimemazine, according to parents’ hypnotic use

Table 4 shows that the association with dispensed alimemazine is stronger for benzodiazepine hypnotics than for z-hypnotics used by parents. In addition, the table shows that the number of prescriptions for hypnotics to parents does not alter the likelihood of children receiving alimemazine.

Table 4

The OR with 95% CI (95% CI) for boys and girls 0–3 years of age to be prescribed alimemazine, according to mothers and fathers benzodiazepine and z-hypnotic use and parents number of hypnotic prescriptions


We confirmed our hypothesis that parents who had previously used hypnotics had an increased likelihood of their child receiving a prescription for alimemazine. If the mother had been dispensed a prescription for hypnotics, the chance that her child would receive alimemazine doubled.

There could be several different reasons why we found an association between parents’ and children's hypnotics prescriptions. One explanation is that biological traits are inherited. This could be that sleep prolems as such are inherited24 ,25 or problematic sleep related to conditions such as anxiety and depression.26

The association could also be explained by parents’ tolerance for their child's sleep problem. Parents who previously struggled with sleep could be more likely to perceive their child's lack of sleep as problematic. Hypnotics users more often struggle with physical and mental health problems.27 This could render parenting and tackling sleep problems more challenging. The chance of a child receiving hypnotics was even bigger when the mother had used benzodiazepine hypnotics. Using benzodiazepine instead of z-hypnotics is related to psychiatric comorbidity.28 In cross-sectional studies an association between children's measured sleep problems and maternal mental health has been found.29–31 We found that mother's previous antidepressants use had an independent impact on alimemazine use in children; although the mental health of the mother might have changed since before the pregnancy, this could indicate that alimemazine prescribed to infants/toddlers is more a sign of a tired parent rather than of a sick child. The finding that the non-antibiotic users were more likely than users to receive alimemazine when both parents had used hypnotics could support this hypothesis.

The effect of parents use of hypnotics on dispensed alimemazine to children could also be explained by parents attitudes or expectations towards health and drugs. A Dutch study found that children of parents who used more drugs, including psychotropic drugs, more often were recurrent user of antibiotics.32 One might imagine that parents who frequently visit the doctor, also due to anxiety, more often are prescribed medicines and this could be true for their children as well. Children and young people with filled precriptions for asthma drugs are more likely to have additional diagnoses unrelated to asthma than others.33

An association between dispensed opioids and benzodiazepines/z-hypnotics to mothers and those dispensed to their teenage offspring has been found.19 ,20 This phenomenon could resemble our findings.

The three percent prevalence for alimemazine use during the first 3 years of life found in our study is slightly lower than previous findings in 1988–1990.1 ,34 This suggests that more doctors are aware that alimemazine may not be the best treatment for sleep problems in children. In Italy, UK and the Netherlands alimemazine is not as frequently used as in Norway, but niaprazine is often prescribed to the children aged 0–2 years in Italy.35 Our data show that around 2.5% of parents filled a prescription for hypnotics during a year. In the adult population as a whole the 1-year prevalence of hypnotics is approximately 8.2%.3 The lower percentage of users would be expected in a group of younger adults.

This study offers a complete capture of all dispensed alimemazine to an entire birth cohort of Norwegian children and their parents. The continuous data collection ensure data without recall bias and loss to follow-up. The drugs in question are all prescription-only drugs and only dispensed drugs are registred. As alimemazine is the most widely used hypnotic drug for Norwegian children in the age group 0–3 years3 ,34 our results can be interpreted as hypnotic use in general in this population.

Information on deaths, emigration and purchase of medicines from abroad or over the internet was unfortunately not available. Information on diagnoses was limited. As exposure, we chose to include only benzodiazepines and z-hypnotics as these are the most commonly used hypnotics for adults in Norway. Pregnancy is not included as exposure as the use of dispensed hypnotics reduces during pregnancy36 and women who use hypnotics during pregnancy might represent a more complex group of mothers, not within the scope of our study. We have looked at hypnotic use in a limited time period, thus missing prescriptions previous to and after our exposure period. But by our method we believe we have captured frequent users and we have no reason to believe that our approach could overestimate the effect of parents’ hypnotic use on children’s alimemazine use.

The relatively high prevalence of dispensed alimemazine in the age group 0–3 years in Norway and its relationship with hypnotics dispensed to parents should encourage doctors to address the whole family's sleep/waking habits and expectations before prescribing hypnotics to children.


Thanks to the Medical Research Programme, University of Oslo, the Medical Birth Registry of Norway (MBR), the Norwegian Prescription Registry (NorPD), and colleagues at the Centre of Drug and Addiction Research (SERAF), University of Oslo and at the Department of Pharmacoepidemiology, Norwegian Institute of Public Health.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Contributors IH had the day to day responsibility of the article, applied for data, did the analysis, wrote the first manuscript, coordinated the editing of the article, wrote the revised manuscript and wrote the decision letter and approved the final draft before submitting. MH contributed in choosing and performing the analysis, reviewed and revised the manuscript, contributed in the revised manuscript and the response to the decision letter. She approved the final draft before submitting. SS participated in analysing and choosing, statistics, reviewed and revised the manuscript, contributed in the revised manuscript and the response to the decision letter. She approved the final draft before submitting. JB had the idea and designed the study, supervised the application for data and the analyses, reviewed and revised the manuscript, contributed in the revised manuscript and the response to the decision letter and approved the final draft before submitting.

  • Funding Funded by the Norwegian Research Council through the Medical Research Programme, University of Oslo. Funder had no direct role in the study and no own interests in the result. The Medical Student Research Program has grant number: Norwegian Research Council-163211/F20. The project was also funded by the Norwegian research council grant number: Norwegian Research Council-07/3779-29.

  • Competing interests None.

  • Ethics approval The identification numbers used in this study are encrypted. The researchers have, at no point, had access to information that might expose the identity of individuals such as the exact birth date. The project was approved by the Norwegian Data Inspectorate.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement Data used is owned by the Norwegian Institute of Public Health. Access to the data is either partially possible on the webpage or by application to the Norwegian Prescription Database and the Medical Birth Registry of Norway.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.