Psychostimulant Prescription for ADHD in New South Wales

Abstract

Objective: There is a need for Australian studies of ADHD that utilize the individual child as the unit of analysis because they provide a more accurate picture of national patterns (in new prescriptions, start age, and duration). The aim of this study was to build toward a national picture of patterns in psychostimulant use for ADHD by undertaking a retrospective analysis of archival data on prescriptions within New South Wales (NSW), Australia’s most populated state. Method: A person-based data set was used to assess (a) rate of new prescriptions by age group, (b) demographic characteristics (age of start, male:female ratio), (c) duration of use, and (d) comparisons across the two decades. Results: Five findings were observed: (a) The prevalence of psychostimulant use was 1.24% in 2010, (b) there was significant variability in the rate of new prescriptions by age group after 2003, (c) start age declined over the 1990 to 2000 period, but began to increase from 2000 to 2010, (d) the male:female ratio declined, and (e) the duration of psychostimulant use declined consistently. Conclusion: Results suggest disconnect between persistence across the life span and actual treatment patterns. A decline in medication treatment for more than 1 year and the growing proportion of discontinuous treatment suggests a need for strategies to assist families with the transition onto and off medications.

Keywords

ADHD pharmacotherapy psychostimulants duration

ADHD is defined as a persistent pattern of inattentive or hyperactive-impulsive behavior, which can result in performance issues in social, educational, or work settings (American Psychiatric Association [APA], 2013). Psychostimulant medication is now a commonly recommended treatment for ADHD. The effectiveness of this treatment for ADHD is well documented (Crenshaw, Kavale, Forness, & Reeve, 1999; Greenhill, 2002; Multimodal Treatment Study of Children with ADHD [MTA] Cooperative Group, 2004) and contributed to a pronounced increase in psychostimulant use for ADHD in Western nations over the last two decades (Banaschewski, Coghill, & Santosh, 2006; Berbatis, Sutherland, & Bulsara, 2002; Findling, 2008; Olfson, & Marcus, 2009; Prosser & Reid, 2009; Stephenson, Karanges, & McGregor, 2013; Visser, Lesesne, & Perou, 2007; Zito, Safer, & DeJong-VanDenBerg, 2008).

ADHD diagnoses increased dramatically in Western nations throughout the 1990s and early 2000s (Berbatis et al., 2002; Conners, 2000; Knopf, Holling, Huss, & Schlack, 2011; Paterson, 2013; Robison, Sclar, Skaer, & Galin, 1999; Safer & Krager, 1988; Swanson, Lerner, & Williams, 1995; Visser et al., 2014). In part, the increase was due to the fact that ADHD was likely underdiagnosed previously. However, the increases in ADHD diagnoses and medication treatment have led to controversy—see Hinshaw and Scheffler (2014) for an overview. Estimates of levels of ADHD diagnosis and drug treatment vary greatly. For instance, an international review of 39 studies estimated ADHD prevalence between 2% and 18% (Skounti & Philalithis, 2007), while another study estimated worldwide prevalence to be 5.3% (Polanczyk, DeLima, Horta, Biederman, & Rohde, 2007). Historically, North American estimates of prevalence have been as low as 0.4% (Sherman & Hertzig, 1991) and as high as 23% (Shaywitz & Shaywitz, 1988), although more recently, consensus has converged around prevalence rates of 2% to 9% (Center for Disease Control and Prevention, 2005; Perwein, Hall, Swensen, & Swindle, 2004; Visser, Bitsko, Danielson, & Perou, 2010).

In other nations using the APA’s (2013) Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5), the prevalence of ADHD diagnosis is also estimated between 2% and 9% (Akinbambi, Liu, Pastor, & Reuben, 2011; Skounti & Philalithis, 2007). Germany and the United Kingdom report much lower estimates of prevalence (Knopf et al., 2011; McCarthy et al., 2012), while Australia, New Zealand, and Canada report levels similar to the United States (Berbatis et al., 2002). Prevalence estimates within Australia have ranged from 1.6% (Valentine, Zubrick, & Sly, 1996) to 11% (Sawyer et al., 2000). Such estimates of prevalence in Australia are based either on overall consumption of psychostimulants (which underestimate ADHD) or community studies (which overestimate ADHD). As we have noted previously (Prosser & Reid, 1999), there is a need for Australia-wide studies that utilize epidemiological methods with the individual child as the unit of analysis. This is because such analysis provides a more accurate picture of national patterns (in new prescriptions, start age, and duration) and allows for interesting regional comparisons. While we use stimulant prescription as a proxy for ADHD diagnosis in Australia, the identification of such patterns can also enable more detailed consideration of clinical implications and inform clinical behavior. However, due to each Australian jurisdiction independently recording psychostimulant authorizations for ADHD (and inconsistencies in this recording and the public availability of this data), such a national view is not easily accessible. In this article, we take a further step toward an Australia-wide understanding by replicating the methods of a previous study in another jurisdiction.

In Australia, the rate of dispensation of psychotropic drugs (excluding ADHD medications) has remained relatively constant (Stephenson et al., 2013). However, the average growth rate in children being medicated for ADHD in Australia was 4.7% per year between 2002 and 2009, which compared with 2% per year in the United States over a similar period (Stephenson et al., 2013). One area with which it is difficult to distil consistent estimates of patterns in use is the rate of new ADHD prescriptions both within and between nations. Between 1994 and 2000, psychostimulant consumption for ADHD grew at a rate of 12% per year across 10 countries (Berbatis et al., 2002), and between 1995 and 2000 the number of youth-prescribed medication for ADHD doubled in the United States (Safer & Krager, 1988). Recently, it has been estimated that approximately 4.3% of North American children are treated with medication for ADHD (Castle, Aubert, Verbrugge, Khalid, & Epstein, 2007; MTA Cooperative Group, 2004; Visser et al., 2007). Rates in Germany and other European nations have also grown but remain significantly lower than U.S. levels (Knopf et al., 2011). However, recent data from Sweden showed a growth rate of 34% per year between 2006 and 2009 (Zetterqvist, Asherson, Halldner, Langstrom, & Larsson, 2013), while in Australia, it has been estimated that new prescriptions for ADHD grew 26% per year between 1984 and 2000 (Berbatis et al., 2002) and by almost 73% between 2000 and 2011 (Paterson, 2013). By reporting on the rate of new prescriptions for ADHD in Australia’s most populated jurisdiction (New South Wales [NSW]) between 1990 and 2010, this article will contribute important additional information about longitudinal trends in the rate of new prescriptions.

Another area that lacks clear data is which children are treated with psychostimulants for ADHD and at what age. International studies over the last decade suggest that while boys continue to be more likely than girls to be medicated for ADHD (Barbaresi et al., 2006; Nussbaum, 2012; Sclar et al., 2012; Visser et al., 2010), the ratio of males to females is slowly declining (Prosser & Reid, 2013; Robison et al., 1999; Visser et al., 2007). Some studies have reported the start age at which medication treatment begins is between 7 and 10 years (Barbaresi et al., 2006; Rappley, Gardiner, Jetton, & Houang, 1995; Zito, Safer, dosReis, Madger, & Riddle, 1997); however, there is still little information on international treatment duration trends (Zetterqvist et al., 2013). Estimated duration of psychostimulant use in North American studies has ranged from 2.5 years (Barbaresi et al., 2006) to 8 years (Schneider & Eisenberg, 2006). Understanding the duration of medication use is important for clinical practice, while there is limited information about the long-term effects of psychostimulant treatment (Barbaresi et al., 2006; Poulton et al., 2013). This article seeks to contribute important information by reporting on trends in gender ratio, start age, and duration of medication use within the NSW context.

It should be noted that there is a sizable body of literature that examines the patterns in psychostimulant treatment for ADHD internationally (Berbatis et al., 2002; Polanczyk et al., 2007; Skounti & Philalithis, 2007) and that there is good evidence of considerable intra-national variability (Berbatis et al., 2002; Bokhari, Mayes, & Scheffler, 2005; Knopf et al., 2011). However, variation within nations over time has gone largely unnoticed. The public release of information on psychostimulant use for ADHD by Australian jurisdictions is inconsistent, with one state publishing data annually and others not having released data for more than a decade. In response, this article examines patterns in psychostimulant use in NSW during the period 1990 to 2010. There has not been a longitudinal study of patterns of psychostimulant use in NSW for over a decade (Salmelainen, 2002), and while this present study uses different methodological approaches, it includes data from the previous study period and extends it a further 10 years. Importantly, the article replicates elements of the methodology of a previous study in South Australia from 1990 to 2006 (Prosser & Reid, 2009) to provide a contemporary point of comparison and thus allows for comparisons across jurisdictions over time.

In this previous article, we reported analysis of psychostimulant use in the city of Adelaide, South Australia. Based on a data set of 7,489 cases, we found that patterns of psychostimulant use had followed U.S. trends, including a rapid initial growth and significant subsequent volatility in prescribing practices. We found more than the 16-year period that:

The male:female ratio declined from 5.4:1 to 4.3:1.

The average start age declined from 9.35 years to 8.84 years.

The average duration of treatment declined from 2.47 years to 2.05 years.

Due to the larger and more complex population in this present study (which covers regional and metropolitan NSW and includes more than 88,000 cases), the geographic and socio-economic analysis of this data set will be the subject of subsequent articles. In this article, we reapply the previous methodology in the NSW context between 1990 and 2010; we hypothesize that the patterns previously found in the South Australian population will be reproduced in that context.

Method

The state of NSW is the major population center of Australia. As of 2010, the population of NSW was 7,221,468 persons. The state of NSW has one major metropolitan area, Sydney (population 4.6 million), while two smaller cities, Newcastle and Wollongong (combined population of approximately 700,000), are the fastest growth areas outside Greater Sydney (Sawyer et al., 2000). The Australian Bureau of Statistics provided population data from the 2011 Census, indicating that the population of 5-to-17-year-olds was 1,161,339 at that time.

Within NSW, data on psychostimulant prescription is recorded in the Pharmaceutical Drugs of Addiction System, which is maintained by Pharmaceutical Services Unit of the Legal and Regulatory Services Branch of the NSW Ministry of Health (MoH). Because psychostimulants are classified as “drugs of dependence,” all medical practitioners in NSW are required to access either a general or individual authorization from the MoH to treat a patient with psychostimulants. General approvals are available to consultant pediatricians and psychiatrists who are members of the Royal Australian and New Zealand College of Psychiatrists. These practitioners do not need an individual approval for prescription provided they follow routine prescribing criteria. All other medical practitioners who wish to prescribe must seek individual approval. This data set represents all children from ages 5 to 17 who have been authorized to receive psychostimulants either by individual or general authorizations. After ethics board approval, data were obtained for 88,204 persons who began medication treatment between 1990 and 2010. The NSW data set records the date of commencement for every authorization, and due to prescriptions being valid for 6 months (unless otherwise instructed by the prescriber), this provides an estimate of duration of use.

Data provided by the MoH included gender, date of birth, and date that each prescription of psychostimulants was initially authorized (from which duration can be calculated). From the data, we calculated the start age (date psychostimulants were initially authorized minus birth date), duration of treatment (date of last authorization [plus 6 months] minus date psychostimulants were authorized), and the number of children beginning psychostimulant treatment per year. Because treatment was often discontinuous (i.e., there were gaps in prescriptions), we identified the children with gaps in treatment. We defined a gap as a period of 1 year or more without an authorization. We computed the average time off medication for each child. For all children, when calculating duration of treatment, we subtracted out the time that children did not have an active authorization. The censor date was May 30, 2012 (i.e., children who were still being treated on that date constitute cases without a known duration of treatment).

We used descriptive statistics to report on yearly prescription rates, demographic information, and duration of use. However, we also expanded our analysis of duration, by addressing discontinuous treatment (i.e., gaps in authorizations), and we improved our analysis of trends by fitting both linear and quadratic regression models to these data. We compared treatment patterns across two decades to determine whether there were differences across 1990 to 1999 when medication use in South Australia was increasing, and 2000 to 2010 when the rate of prescriptions had stabilized to determine whether the same trends were apparent in NSW. In addition, we calculated effect sizes for inferential tests when possible.

Results

Figures 1 to 4 provide an overview of the patterns in psychostimulant use for the period 1990 to 2010. The number of young persons with active authorizations for psychostimulants as of December 31, 2010, was 14,986. This indicates an approximate prevalence rate of children from 5 to 17 years diagnosed with ADHD and medicated of 1.24% in NSW at this time.

Figure 1.

Rate of new cases (per 1,000 population).

Figure 2.

Male to female ratio, 1990 to 2010.

Figure 3.

Change in mean start age, 1990 to 2010.

Figure 4.

Mean treatment duration, 1990 to 2010.

Rate of New Cases

Figure 1 shows the rate of new cases by age group over the period of 1990 to 2010 and highlights a rapid increase from 1990 to 1995. As expected, the highest rate was in the 5-to-9-year age group followed by the 10-to-14-year age group. The rate of new cases per year in the 5-to-9-year group continued to increase until 1998, but then remained above five new cases per thousand until 2000. However, there was considerable variability in the 5-to-9-year group after 2001. By contrast, the number of new cases in the 10-to-14-year group peaked in 1994, declined until 2003, and then closely paralleled the variation in new cases of the 5-to-9-year group until 2010. There was minimal variability in new cases for the 15-to-17-year group, with the rate remaining steady since 1996.

Male to Female Ratio

Overall male to female ratio was 3.96:1 with a 99% confidence interval (CI) ranging from 3.93:1 to 4.00:1. For the first decade, the ratio was 4.54:1 (CI [4.49, 4.59]); for the second decade, the ratio was 3.56:1 (CI [3.52, 3.61]). The difference in ratios between decades was significant, χ²(1) = 206.60, p < .0001, indicating that the likelihood of males (relative to females) being prescribed psychostimulants decreased by 4.8% between decades. Figure 2 shows that the ratio of males to females steadily dropped from 1990 to 2000 then stabilized between 2001 and 2010. The ratio decreased from a high in 1992 of 5.85:1 to a low in 2006 of 3.31:1. Hence, our hypothesis of declining male to female ratio was confirmed. The largest mean ratio was found for the 10-to-14-year-old group (4.15:1) followed closely by the 5-to-9-year-old group (4.05:1) and then the 15-to-17-year-old group (2.77:1). All three groups share the same basic downward and decelerating trend over time.

Mean Start Age

The overall mean start age was 9.81 years (SD = 3.08); the median age was 9.22 years. Start age differed, between decades with children in the first decade beginning psychostimulants at a slightly older age (9.88 vs. 9.75), t(86859.94) = 6.30; however, the effect was very small (g = 0.04), which suggests little practical significance. The respective start ages for males and females were 9.77 (SD = 3.05) and 9.98 (SD = 3.20). This difference was statistically significant, t(26492.93) = 8.37, p < .0001, but once again, the effect size was very small (g = −0.07). The hypothesis of a declining start age was not confirmed. Rather, a visual and statistical analysis indicated that change over time was curvilinear in nature (see Figure 3) with mean start age decreasing between 1990 and 2003 from a high of 10.53 in 1991 to a low of 9.54 in 2003, and then subsequently increasing between 2004 and 2010.

Duration of Treatment

Of the 88,204 cases, 76,139 youth (61,054 males and 15,085 females) were no longer being prescribed psychostimulants. Of these cases, 39,475 were from 1990 to 1999, and 36,664 were from 2000 to 2010. The overall mean treatment duration was 2.73 years (SD = 2.45). The median treatment duration was 1.94 years. The treatment duration for 1990 to 1999 was 3.32 years (SD = 2.75), and the treatment duration for 2000 to 2010 was 2.10 years (SD = 1.88); this difference was statistically significant, t(70116.49) = 71.66, p < .0001, and represents a moderate to large effect size (g = 0.52). A regression analysis revealed a more fine-grained pattern of treatment duration over time, which decreased significantly from 1990 to 2010, and the trend is slightly accelerating over time (see Figure 4). Treatment duration was highest in 1993 at 3.82 years and lowest in 2010 at 0.90 years representing a 73.3% decrease in duration over the 21-year period or a decrease of 44 days per year although this rate of change is not constant as indicated by the curvilinear trend. Hence, our hypothesis in relation to declining durations of medication treatment was confirmed. However, the accelerating trend should be considered in light of discontinuous treatment—some of the children prescribed medication between 2000 and 2010 will receive medication at a later date so these children represent right-censored cases (i.e., cases for which we do not know the true survival time). Therefore, while there is strong downward trend in duration, these data overstate the degree of change over time.

Over the 2 decades, 7.8% of children who were prescribed psychostimulants received treatment for 6 months or less. An additional 20.4% of children received treatment for 1 year or less. These proportions are substantially different between decades. For 1990 to 1999, 4.8% of children received treatment for 6 months or less, and another 15.5% received treatment for 1 year or less. However, for 2000 to 2010, 11.0% of children received treatment for 6 months or less, and an additional 25.6% received treatment for 1 year or less.

In addition, there was a moderate to large relationship between age at first prescription and treatment duration (r = −.28, p < .0001) with children who start treatment at a younger age receiving longer periods of treatment. For example, the difference in treatment duration for children 5 to 9 years and children 10 to 14 years was 1.06 years (g = 0.45), while the difference between children 5 to 9 years and children 15 to 18 years was 1.73 years (g = 0.66). Both of these specific contrasts were statistically significant at the .0001 alpha level.

Discontinuous Treatment

Of the total sample, 20.7% of children experienced discontinuous treatment, that is to say children ceased medication for at least 1 year before resuming treatment. Of the 18,301 children with gaps in treatment, the length of the gaps ranged from 1.00 to 20.89 years with a mean of 3.71 years (SD = 2.41) and a median of 3.03 years. It should be noted that this mean represents the average length per child (with all gaps aggregated together) and not the average length of each gap in treatment. There was no significant relationship between gender and the likelihood of having discontinuous treatment, χ²(1) = 0.03, p = .88. However, there was a significant difference in the mean treatment duration for children with discontinuous treatment (M = 3.80 years, SD = 2.39) compared with children with continuous treatment (M = 2.47 years, SD = 2.40), where children with discontinuous treatment received significantly longer treatment duration, t(22835.43) = 60.95, p < .0001, g = 0.55. There was also a significant difference in the proportion of children who experienced discontinuous treatment between the two decades, χ²(1) = 755.85, p < .0001, with the rate of discontinuous treatment for 1990 to 1999 (24.8%) being 43.4% higher than 2000 to 2010 (17.3%).

Discussion

The present results show that the patterns of psychostimulant treatment of ADHD for children in NSW are similar to previous international and Australian studies, although there are some notable differences. A comparison between these present findings and our previous South Australian study is provided in Table 1.

Table 1.

Comparison Between General Findings (NSW and SA).

Finding	SA 1990-2000	SA 2001-2006	NSW 2001-2010
New prescribe	Sharp increase (1990-1995) Steady decline (1995-2000)	Sharp increase (2000-2004) Sharp decline (2005-2006)	Brief decline (2002) Steady increase (2002-2004) Steady decline (2004-2006) Steady increase (2006-2010)
Male:female ratio	5.4:1	4.3:1	3.6:1
Mean start age (years)	9.35	8.84	9.56
Mean duration (years)	2.47	2.05	3.09 (2001-2006) 2.06 (2007-2010)

Note. NSW = New South Wales; SA = South Australia.

The 2010 prevalence of psychostimulant use was 1.24%, which is below the frequently estimated range of 2% to 9% (Center for Disease Control and Prevention, 2005; Perwein et al., 2004; Visser et al., 2010) but is a slight increase over the last major estimate of psychostimulant consumption in 2000 (Berbatis et al., 2002). The rapid overall increase in psychostimulant use from 1990 to 1995 is consistent with previous studies in Australia (Berbatis et al., 2002; Prosser & Reid, 2009; Valentine et al., 1996) and mirrors a similar trend in the United States (Akinbambi et al., 2011; Safer, Zito, & Fine, 1996). Notably, the rate of new cases in the 15-to-17-year age group remained relatively stable after 1995, which also aligns with our previous Australian study but contradicts findings of growing prevalence in the 15-to-20-year age group in North America (Castle et al., 2007), Sweden (Zetterqvist et al., 2013), and Great Britain (Wong et al., 2009). In these countries, higher levels may be associated with the improved international identification of ADHD within persons older than 18 years (Castle et al., 2007). However, a recent study in NSW indicated less scope for improvement in this area as the prescribing practices of child and adult psychiatrists (often cited as a cause of different diagnostic rates) have been shown to be more consistent in that context (Mitchell et al., 2012). Such differences in the broader Australian context are worthy of further investigation.

Also requiring further consideration is the variations in the rate of new cases for the 5-to-9-year and 10-to-14-year age groups since 2003. These fluctuations mirrored each other so closely it is noteworthy. The authors contacted the MoH to inquire whether any regulatory changes or refinements to data collection practices occurred during this period that might provide explanation, but none were identified. It has been noted previously that within smaller Australian jurisdictions, the arrival or departure of a medical specialist (Paterson, 2013; Prosser & Reid, 2013), individual prescribing practices (Preen, Calver, Sanfilippo, Bulsara, & Holman, 2008), or prominent media reports and public events (Prosser & Reid, 2009) can influence annual rates of new cases. Although the authors thought this unlikely given the size of the NSW data set, we again contacted the MoH, which reported no such changes in the years around 2004 and 2007. Hence, we have found no obvious explanation for this unexpected mirroring in fluctuations. Interestingly, our previous South Australian study also found a peak in new cases in 2003 to 2004 and a sharp decline by 2006.

In the past, lower levels of diagnosis of ADHD among girls have been well documented (Nussbaum, 2012). Meanwhile, the growing number of girls being medicated for the disorder over the last two decades has been associated with the overall increasing trend in psychostimulant use for ADHD (McCarthy et al., 2012; Sclar et al., 2012). This present study found that more males were likely to be treated with psychostimulants, but that this did not prevent a decrease in the ratio of boys to girls being medicated for the disorder. Within the NSW data, an overall trend toward a declining male:female ratio was identified, which concurs with earlier Australian studies (Preen, Calver, Sanfilippo, Bulsara, & Holman, 2007; Prosser & Reid, 2009). This decline has stabilized since 2001, which may indicate that previous concerns about under-reporting (McCarthy et al., 2012) and the suitability of diagnostic criteria developed with boys being used with girls (Nussbaum, 2012) are beginning to be addressed.

Another interesting aspect of the data was the trends in start age of psychostimulant use. Our previous South Australian study identified a downward trend in mean start age in the range of 6 months between 1990 and 2006, due primarily to an increase of boys being medicated at younger ages. However, it should be noted that a small increase was recorded after 2004 in that study. This present study identified a similar decline from 1990, but then a significant increase in mean start age between 2004 and 2010. The decline in mean start age in the first decade is due to there being a greater proportion of children within the 5-to-9-year group. However, the later increase was produced by more psychostimulant treatment among children who began medication at 15 to 17 years. It is possible that a similar trend may be present in the more recent South Australian data post 2006.

A surprising aspect of the data was the downward trend in duration of medication use and particularly the decrease in numbers of young people using medication for longer than 1 year. However, this aligns with international studies that indicate that up to 50% of children cease medication use within 12 months (Bussing et al., 2005; van den Ban et al., 2010). Almost 30% of children in this present study received medical treatment for less than 1 year. It should be emphasized, though, that the prevalence of discontinuous treatment presents a significant barrier to an analysis of mean treatment duration over time. This is because children who appear to have ceased treatment in recent years may resume treatment eventually, thus increasing their duration and bringing it more in line with the children from the earlier years of the study. That more than 20% of children had gaps in drug treatment and that the average of the total time off treatment for each of these children was around 3 years suggests caution around using medication as the only form of treatment for ADHD.

Limitations

There are several challenges inherent for any study that assesses patterns in psychostimulant use across Australia. These include the following:

Australia is a federation of autonomous states and territories with individual jurisdictions having different authorization requirements for prescribing and different systems for recording these authorizations (Swanson et al., 1995);

Previous studies comparing prescription patterns between jurisdictions (Swanson et al., 1995; Valentine et al., 1996) are inconclusive due to large gaps in the data in and between jurisdictions; and

National studies use international narcotics control records to report levels of psychostimulant consumption and to produce Australian estimates of prevalence (Berbatis et al., 2002), but these are difficult to relate to specific jurisdictions and unable to be linked to patient demographic data.

This study of one Australian jurisdiction shares these limitations, and hence the authors warn against broad generalization based on its findings.

There are also limitations specific to the NSW data set. First, prior to 1995, only individual approvals were in place, while from 1996 onward, both individual and general approvals were available. This means that data prior to 1996 is not strictly comparable with subsequent NSW data. Second, while medical practitioners are required to report on prescribing on a monthly basis, they do not always do so, which can result in missing data on cessation date. On these occasions, a date 6 months after the prescription date was inserted into the data set, but this estimate may not reflect that actual full duration of the data set. Third, the information that has been accessed is only of prescriptions written under approval, we do not know whether prescriptions were filled. However, it can be assumed that non-filled prescriptions will not be renewed and contribute only a maximum of 6 months to estimates of duration. Finally, it should be noted that the censor date for this data set was in 2012, while the last start date for analysis was in December 2010. As a result, only the patients who had ceased treatment by the censor date were included in calculations of duration. As the likelihood of restarting treatment after 1-year cessation is significantly higher between 1990 and 1999 compared with 2000 to 2001, it would stand to reason that the duration estimate for the second decade might be biased due to right censoring of data.

Clinical implications

This data on patterns in psychostimulant use for ADHD in NSW across two decades suggest a number of clinical implications. First, the start age of children is increasing. This aligns with a broader trend embodied in the new DSM-5 (APA, 2013) to recognize a later age of onset for ADHD. This increase may result in more demand for diagnostic assessments, may require greater awareness of the complex interrelation of ADHD and early puberty, and may involve more complex and time-consuming assessments for clinicians (Prosser & Reid, 2013). Given the recognized shortage of child psychiatrists in some jurisdictions (Paterson, 2013), this growing demand may also result in greater involvement of pediatricians in diagnosis and result in broader variations in future diagnostic practice and patterns of use (Mitchell et al., 2012). Furthermore, assessment and medication at older ages will inevitably involve clinicians in liaison with a wider range of education professionals in more complex secondary school settings, which again can be time-consuming and will require establishing clear lines of communication.

The present study found that almost 30% of children were receiving medication treatment for less than 1 year. With this in mind, clinicians will have an important role in supporting the accessibility of other non-clinical forms of support as part of a multi-modal treatment approach. Furthermore, the duration of psychostimulant treatment has decreased consistently across two decades, which stresses the need for adequate resourcing to expand the accessibility of non-medication supports.

The study also found that the proportion of children having discontinuous treatment grew over the two decades. In the light of the identified increase in mean start age, one possible explanation for this trend is the use of medication around times of greater social and academic challenge, such as transitioning into secondary school or study for senior exams. Such strategic use of medication has been reported within qualitative studies (Prosser, 2006) and is worthy of further empirical examination. Together, these findings around shorter continuous psychostimulant use for ADHD provide a context into which to place concerns in relation to the negative health effects of long-term medication use for ADHD (Poulton et al., 2013). They also highlight the clinical importance of the inclusion of strategies to assist families with the transition onto and off medications.

In summary, this retrospective analysis of psychostimulant prescription data for children with ADHD in NSW found two key trends that have continued across the two decades between 1990 and 2010. First, the nature of the treated population and the patterns of that treatment have changed markedly. Second, the trend in duration of medication use suggests that there appears to be a disconnect between what we know about ADHD in terms of persistence across life span and actual treatment patterns. Our evidence suggests that medication is not a viable long-term treatment approach for the majority of children and youth in this study. We recommend further research into the presence of similar trends in other Australian jurisdictions as well as into their clinical implications.

Footnotes

Acknowledgements

The authors acknowledge the support of the Pharmaceutical Services Unit, Legal and Regulatory Services Branch of the New South Wales Ministry of Health and particularly that of its Senior Analyst Information Management, Pia Salmelainen.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Brenton Prosser is Senior Research Fellow in the Centre for Research and Action in Public Health at the University of Canberra, Australia. He is also adjunct to the School of Sociology at the Australian National University. Dr. Prosser received his Doctor of Philosophy from the Flinders University of South Australia, where his thesis focused on the prevalence of ADHD in Australia, along with its professional and policy implications.

Matthew C. Lambert is an early career scholar who holds an appointment as an assistant professor of research methodology in the Barkley Center at the University of Nebraska–Lincoln. His research primarily centers on psychological and educational measurement, and intervention effectiveness for children’s mental health services. He is an author of the Behavioral and Emotional Ratings Scale–3, a strength-based assessment that is used as a service planning and progress monitoring tool in mental health and school settings.

Robert Reid, PhD, is a professor in the Department of Special Education and Communication Disorders at the University of Nebraska–Lincoln. He received his doctorate in special education from the University of Maryland in 1991. Dr. Reid’s co-developed the ADHD-IV Rating Scale (1998).

References

Akinbambi

L. J.

Liu

Pastor

P. N.

Reuben

C. A.

(2011, August). Attention deficit hyperactivity disorder among children 5-17 years in the United States, 1998-2009 (NCHS Data Brief, 70). Hyattsville, MD: Centers for Disease Control and Prevention, U.S. Department of Health and Human Services.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.

Banaschewski

Coghill

Santosh

(2006). Long-acting medications for the hyperkinetic disorders: A systematic review and European treatment guideline. European Child & Adolescent Psychiatry, 15, 476-495.

Barbaresi

W. J.

Katusic

S. K.

Colligan

R. C.

Weaver

A. L.

Leibson

C. L.

Jacobsen

S. J.

(2006). Long-term stimulant medication treatment of attention-deficit/hyperactivity disorder: Results from a population-based study. Journal of Developmental and Behavioral Pediatrics, 27, 1-10.

Berbatis

C. G.

Sutherland

V. B.

Bulsara

(2002). Licit psychostimulant consumption in Australia, 1984-2000. Medical Journal of Australia, 177, 539-543.

Bokhari

Mayes

Scheffler

(2005). An analysis of the significant variation in psychostimulant use across the U.S. Pharmacoepidemiology and Drug Safety, 14, 267-275.

Bussing

Zima

Mason

Hou

Wilson

G. C.

Forness

(2005). Use and persistence of pharmacology for elementary schools students with attention-deficit/hyperactivity disorder. Journal of Child and Adolescent Psychopharmacology, 15, 78-87.

Castle

Aubert

R. E.

Verbrugge

R. R.

Khalid

Epstein

R. S.

(2007). Trends in medication treatment for ADHD. Journal of Attention Disorders, 10, 335-342.

Center for Disease Control and Prevention. (2005). Prevalence of diagnosis and medical treatment for attention-deficit/hyperactivity disorder. The Journal of the American Medical Association, 294, 2293-2295.

10.

Conners

C. K.

(2000). Attention-deficit/hyperactivity disorder: Historical development and overview. Journal of Attention Disorders, 3, 173-191.

11.

Crenshaw

T. M.

Kavale

K. A.

Forness

S. R.

Reeve

R. E.

(1999). Attention deficit hyperactivity disorder and the efficacy of stimulant medication: A meta-analysis. In Scruggs

Mastropierie

(Eds.), Advances in learning and behavioral difficulties (pp. 135-165). Greenwich, CT: JAI Press.

12.

Findling

R. L.

(2008). Evolution of the treatment of attention-deficit/hyperactivity disorder in children: A review. Clinical Therapeutics, 30, 942-957.

13.

Greenhill

L. L.

(2002). Practice parameters for the use of stimulant medication in the treatment of children, adolescents and adults. Journal of the American Academy of Child & Adolescent Psychiatry, 41, 26S-49S.

14.

Hinshaw

Scheffler

(2014). The ADHD explosion. New York, NY: Oxford University Press.

15.

Knopf

Holling

Huss

Schlack

(2011). Prevalence, determinants and spectrum of attention-deficit hyperactivity disorder (ADHD) medication of children and adolescents in Germany: Results of the German Health Interview and Examination Survey (KiGGS). BMJ Open, 2, 1-12.

16.

McCarthy

Wilton

Murray

M. L.

Hodgkins

Asherton

Wong

I. C. K.

(2012). The epidemiology of pharmacologically treated attention deficit hyperactivity disorder (ADHD) in children, adolescents and adults in UK primary care. BMC Pediatrics, 12(78), 1-11.

17.

Mitchell

P. B.

Levy

Hadzi-Pavlovic

Concannon

P. E.

Hutchins

Mulcahy

Hughes

C. F.

(2012). Practitioner characteristics and the treatment of children and adolescent with attention deficit hyperactivity disorder. Journal of Paediatrics and Child Health, 48, 483-489.

18.

MTA Cooperative Group. (2004). National Institute of Mental Health Multimodal Treatment Study of ADHD follow up: Changes in effectiveness and growth after treatment. Pediatrics, 113, 762-770.

19.

Nussbaum

N. L.

(2012). ADHD and female specific concerns: A review of the literature and clinical implications. Journal of Attention Disorders, 16, 87-100.

20.

Olfson

Marcus

S. C.

(2009). National patterns in antidepressant medication treatment. Archives of General Psychiatry, 66, 848-856.

21.

Paterson

(2013). ADHD in Western Australia: An iatrogenic “epidemic” cured by bureaucracy? Time to focus on the facts. Medical Journal of Australia, 47, 88-89.

22.

Perwein

A. R.

Hall

Swensen

Swindle

(2004). Stimulant treatment patterns and compliance in children and adults with newly treated attention-deficit/hyperactivity disorder. Journal of Managed Care Pharmacy, 10, 122-129.

23.

Polanczyk

DeLima

M. S.

Horta

B. L.

Biederman

Rohde

L. A.

(2007). The worldwide prevalence of ADHD: A systematic review and meta-regression analysis. The American Journal of Psychiatry, 164, 942-948.

24.

Poulton

A. S.

Melzer

Tait

P. R.

Garanett

S. P.

Cowell

C. T.

Baur

L. A.

Clarke

(2013). Growth and pubertal development of adolescent boys on stimulant medication for attention deficit hyperactivity disorder. Medical Journal of Australia, 198, 29-32.

25.

Preen

D. B.

Calver

Sanfilippo

F. M.

Bulsara

Holman

C. D.

(2007). Patterns of psychostimulant prescribing to children with ADHD in Western Australia: Variations in age, gender, medication type and dose prescribed. Australian and New Zealand Journal of Public Health, 31, 120-126.

26.

Preen

D. B.

Calver

Sanfilippo

F. M.

Bulsara

Holman

C. D.

(2008). Prescribing of psychostimulant medication for attention deficit hyperactivity disorder in children: Differences between clinical specialties. The Medical Journal of Australia, 188, 337-339.

27.

Prosser

(2006). ADHD: Who’s failing who? Sydney, Australia: Finch.

28.

Prosser

Reid

(1999). Psychostimulant use for children with ADHD in Australia. Journal of Emotional and Behavioral Disorders, 7, 110-117.

29.

Prosser

Reid

(2009). Changes in use of psychostimulant medication for ADHD in South Australia (1990-2006). Australian & New Zealand Journal of Psychiatry, 43, 340-347.

30.

Prosser

Reid

(2013). The DSM-V changes and ADHD: More than a tweak of terms. Australian & New Zealand Journal of Psychiatry, 47, 1196-1203.

31.

Rappley

M. D.

Gardiner

J. C.

Jetton

J. R.

Houang

R. T.

(1995). The use of methylphenidate in Michigan. Archives of Pediatrics & Adolescent Medicine, 149, 675-679.

32.

Robison

Sclar

Skaer

Galin

(1999). National trends in the prevalence of attention-deficit hyperactivity disorder and the prescribing of methylphenidate among school-age children. Clinical Pediatrics, 38, 209-217.

33.

Safer

D. J.

Krager

J. M.

(1988). A survey of medication treatment for hyperactive/inattentive students. The Journal of the American Medical Association, 260, 2256-2258.

34.

Safer

D. J.

Zito

J. M.

Fine

E. M.

(1996). Increased methylphenidate usage for attention deficit disorder in the 1990s. Pediatrics, 98, 1084-1088.

35.

Salmelainen

(2002). Trends in prescribing of stimulant medication for the treatment of attention deficit hyperactivity disorder in children and adolescents in NSW. Sydney, Australia: NSW Department of Health.

36.

Sawyer

M. G.

Arney

F. M.

Baghurst

P. A.

Clark

J. J.

Graetz

B. W.

Kosky

R. J.

. . . Zubrick

S. R.

(2000). Mental health of young people in Australia. Canberra, Australia: Commonwealth Department of Health and Aged Care.

37.

Schneider

Eisenberg

(2006). Who receives a diagnosis of attention-deficit/hyperactivity disorder in the United States elementary school population? Pediatrics, 117, e601-e609.

38.

Sclar

D. A.

Robison

L. M.

Bowen

K. A.

Schmidt

J. M.

Castillo

L. V.

Oganov

A. M.

(2012). Attention-deficit/hyperactivity disorder among children and adolescents in the United States: Trend in diagnosis and use of pharmacotherapy by gender. Clinical Pediatrics, 51, 584-589.

39.

Shaywitz

S. E.

Shaywitz

B. A.

(1988). Attention deficit disorder: Current perspectives. In Kavanagh

J. G.

Truss

T. J.

(Eds.), Learning disabilities: Proceedings of the national conference (pp. 69-523). Larchmont, NY: Mary Ann Liebert.

40.

Sherman

Hertzig

M. E.

(1991). Prescribing practices of Ritalin: The Suffolk County, New York Study. In Greenhill

L. L.

Osman

B. B.

(Eds.), Ritalin: Theory and patient management (pp. 187-194). Larchmont, NY: Mary Ann Liebert.

41.

Skounti

Philalithis

(2007). Variations in prevalence of attention deficit hyperactivity disorder worldwide. European Journal of Pediatrics, 166, 117-123.

42.

Stephenson

C. P.

Karanges

McGregor

I. S.

(2013). Trends in the utilization of psychotropic medications in Australia from 2000 to 2011. Australian & New Zealand Journal of Psychiatry, 47, 74-87.

43.

Swanson

J. M.

Lerner

Williams

(1995). More frequent diagnosis of attention deficit-hyperactivity disorder. The New England Journal of Medicine, 333, Article 9444.

44.

Valentine

Zubrick

Sly

(1996). National trends in the use of stimulant medication for attention deficit hyperactivity disorder. Journal of Pediatric Child Health, 32, 223-227.

45.

van den Ban

Souverein

P. C.

Swaab

van Engeland

Egberts

T. C. G.

Heerdink

E. R.

(2010). Less continuation of ADHD drug use since the availability of long-acting ADHD medication in children, adolescents and adults under the age of 45 years in the Netherlands. Attention Deficit and Hyperactivity Disorders, 2, 213-220.

46.

Visser

S. N.

Bitsko

R. H.

Danielson

M. L.

Perou

(2010). Increasing prevalence of parent-reported attention-deficit/hyperactivity disorder among children—United States, 2003 and 2007. Morbidity and Mortality Weekly Report, 59, 1439-1443.

47.

Visser

S. N.

Danielson

M. L.

Bitsko

R. H.

Holbrook

J. R.

Kogan

M. D.

Ghandour

R. M.

. . . Blumberg

S. J.

(2014). Trends in the parent-report of health care provider-diagnosed and medicated attention-deficit/hyperactivity disorder: United States, 2003-2011. Journal of the American Academy of Child & Adolescent Psychiatry, 53, 34-46.

48.

Visser

S. N.

Lesesne

C. A.

Perou

(2007). National estimates and factors associated with medication treatment for childhood attention-deficit/hyperactivity disorder. Pediatrics, 119, S99-S106.

49.

Wong

Asherson

Bilbow

Clifford

Coghill

DeSoysa

. . . Taylor

(2009). Cessation of attention deficit hyperactivity disorder drugs in the young (CADDY)—A pharmacoepidemiological and qualitative study. Health Technology Assessment, 13(50), 1-58.

50.

Zetterqvist

Asherson

Halldner

Langstrom

Larsson

(2013). Stimulant and non-stimulant attention deficit/hyperactivity disorder drug use: Total population study of trends and discontinuation patterns 2006-2009. Acta Psychiatrica Scandinavica, 128, 70-77.

51.

Zito

J. M.

Safer

D. J.

DeJong-VanDenBerg

L. T.

(2008). A three-country comparison of psychotropic medication prevalence in youth. Child and Adolescent Psychiatry and Mental Health, 2, Article 26.

52.

Zito

J. M.

Safer

D. J.

dosReis

Madger

L. S.

Riddle

M. A.

(1997). Methylphenidate patterns among Medicaid youth. Psychopharmacology Bulletin, 33, 143-147.