Introduction

Attention-deficit/hyperactivity disorder (ADHD) is characterized by symptoms of inattention and/or hyperactivity and impulsivity that are inappropriate for the individual’s age and developmental level [2]. With an estimated worldwide prevalence of 5.9% [3], it is one of the most commonly diagnosed mental disorders in children.

Reported rates of ADHD vary considerably between studies from different countries [3], and even regionally, within countries [4,5,6,7]. In the case of regional differences, the observed variation cannot be caused by study methodology as comparisons are based on official numbers of registered diagnoses. In Norway, a recent study has empirically demonstrated that the large geographic variation in diagnosis rates in the country cannot be explained by a correspondingly varying symptom load in the population [8]. Likewise, despite close similarities between ADHD treatment guidelines internationally, medication rates also differ markedly both between and within countries [5,6,7, 9,10,11].

ADHD symptoms exist on a continuum, with no absolute cutoff between clinical diagnosis and “normal” functioning [12, 13]. Moreover, a wide range of conditions can have manifestations resembling ADHD symptoms [14,15,16]. As for other mental disorders, there are no biomarkers or decisive tests available for objectively confirming an ADHD diagnosis in clinical practice. Observed symptoms must thus be assessed by a clinician in terms of their onset, course, severity, and resulting functional impairment to decide whether diagnostic criteria are fulfilled and if an underlying vulnerability is the most likely cause. In cases bordering on the threshold for ADHD diagnosis, the diagnostic decision will to some extent be a result of the individual clinician’s judgment.

Unnecessarily receiving a diagnosis and medication or being deprived of an appropriate diagnosis and possibly beneficial treatment both have potential detrimental effects in a life course perspective [17,18,19,20]. The long-lasting debate in the research literature and among practicing clinicians demonstrates concerns over both alternatives, with experts emphasizing different aspects of this dilemma [13]. Central points of concern regarding excessive diagnosis include the various effects of stigma and labeling [17, 18], as well as the general societal medicalization of childhood. Apprehensions associated with overtreatment span from worries about the immediate side effects of medications to the current lack of knowledge about the long-term consequences of their use [20]. On the other hand, receiving a formal diagnosis and treatment may reduce negative consequences of ADHD symptoms [19], as it offers a framework for increased understanding of the individual’s challenges and provides access to medication as a means of potential symptom relief [17]. How clinicians balance these considerations will presumably influence their clinical decisions, which may range from taking a restrictive stance aiming to minimize the use of diagnoses and/or medication, to being more liberal in terms of diagnosing or recommending medication even in less clear-cut cases.

Variation in medical practice, both at the individual and group level (i.e., local practice cultures), is a well-documented phenomenon [21]. Some previous research has examined different aspects of clinicians’ attitudes toward ADHD and how this may influence clinical practice [see e.g., 22–27], though mostly targeting physicians, often in the role of gatekeepers to specialized mental health services. In Norway, ADHD diagnosis and treatment are decided by clinicians working in a specialized and interdisciplinary healthcare setting, where the therapist in charge will often be a clinical psychologist.

The objective of the current study was to describe variation in attitudes toward diagnosis and medication of ADHD among clinicians working in the Norwegian child and adolescent mental health services (CAMHS) outpatient clinics. Further, we wanted to explore if differences in attitudes between clinicians were related to professional background or workplace (clinic). We hypothesized that attitudes vary on a spectrum from “restrictive” to “liberal”. We proposed that such differences might influence the management of children on the margin of ADHD diagnosis, and might help explain the geographical variance in rates of diagnosis and medication of ADHD.

Methods

Survey development

To investigate our objective, we developed a survey aimed at measuring clinicians’ attitudes toward diagnosis and medication of ADHD as latent constructs. A team consisting of two clinical psychologists (AM, IL) and two psychiatrists with extensive clinical experience with ADHD patients (IB, AH) designed the questionnaire. To further improve validity and reliability of the items, five external clinicians with special interest in ADHD tested the survey during its development and were interviewed to collect feedback on points of improvement. In addition, the survey was piloted in one clinic.

Survey design

The survey was developed as a one-time, self-administered, web-based questionnaire. Questions were closed-ended with four or six Likert-scaled, forced choice options (i.e., no neutral response possible). Items were developed assuming variation from restrictive to liberal attitudes toward ADHD diagnosis and medication.

Comment boxes for optional supplementary comments were provided for each item. All except for two background questions were possible to skip. To maximize response rate, we aimed for a survey that was quick to complete (< 10 min). The survey also included some items as part of a broader investigation of ADHD practice in the clinics that are not relevant to the objective in this article and thus not presented here. The complete, translated survey can be found in Appendix A.

To explore variation in clinician attitudes, eight items were used in analyses. To ease communication of results, all items have been given labels that are indicated in italics throughout the text. Three survey items covered background information about respondents (profession/educational background, work experience, and frequency of contact with ADHD patients, Table 1). Three items involved attitudes regarding diagnosis of ADHD (certainty when diagnosing, hypothetical prevalence in an ideal world, Table 2; await making a diagnostic decision, Table 3). One item considered treatment of ADHD (statements about medication, Table 4), and one item was originally intended to cover both diagnosis and medication (over/undertreatment in Norway today, Table 2). Two of these items (await, medication) consisted of a question followed by several statements (“sub-items”) to which respondents indicated to which extent they agreed. See Tables 1, 2, 3 and 4 for wording and options.

Table 1 Sample characteristics
Full size table
Table 2 Items covering attitudes concerning thresholds for diagnosis and treatment
Full size table
Table 3 Item with sub-items covering attitude toward diagnosis (DA)
Full size table
Table 4 Item with sub-items covering attitude toward medication (MA)
Full size table

Respondents

We aimed to include the whole population of clinicians, including all healthcare professionals of various educational backgrounds, currently involved in diagnosing and treating ADHD in the Norwegian CAMHS outpatient clinics. This represents most clinicians working with ADHD among children and adolescents in Norway, as the private health sector in child psychiatry is negligible.

Data collection procedure

As there is no central registry of employees in the 88 CAMHS outpatient clinics in Norway, we approached the heads of all the clinics and asked them to forward the survey to relevant employees in their unit. Each clinic received a unique link, enabling us to give feedback to the clinics on their average scores compared to national data (provided a response rate of > 50%).

The survey was accessible from September to December 2020. A reminder was sent to all heads of clinics two weeks after initial distribution. Clinics that still had not returned any responses a few weeks after the reminder received personalized follow-up emails or a call.

Statistical modeling and analysis

Answers to survey items concerning attitudes were scored along the hypothesized dimension with low scores representing restrictive and high scores liberal attitudes. To estimate response rate, we used the coverage of clinicians per 1,000 children in some catchment areas (made available to us by coauthor IB), which should be roughly generalizable to other regions.

Structural equation modeling (SEM) was applied to estimate and examine the relationships between our hypothesized attitude variables. A two-factor confirmatory factor analysis (CFA) was used to estimate the latent constructs of clinicians’ attitude toward diagnosis (DA) and attitude toward medication (MA).

Three survey items concerning attitude toward diagnosis were excluded from the final DA model, one due to high amount of missing (D8/other; Table 3) and two because of low factor loadings (certainty: \(\lambda\) = 0.12; ideal: \(\lambda\)= 0.11; Table 2). DA was thus estimated by seven sub-items of the item await (D1–D7; Table 3). Residuals were allowed to correlate between some indicators measuring highly related concepts: both D2/trauma and D3/health problems in the family could also be characterized as D1/psychosocial challenges; and D6 and D7 both refer to neurodevelopmental conditions (Table 3).

MA was estimated by seven sub-items of the item medication (M1–M9; Table4). Two sub-items were excluded due to low factor loadings (M2: \(\lambda\) = 0.18; M4: \(\lambda\) = 0.24). Residual correlations were allowed between M3 and M5 (both relating to concerns over side effects).

The item over/undertreatment (Table 2) was not included in the SEM analysis as its wording in terms of both question and options made it unclear where to fit in the model.

Standardized factor scores for DA and MA were extracted from the two-factor CFA model. The extent of variation in attitudes attributable to profession and clinic level was examined by intraclass correlation coefficients (ICC) in variance-components models, based on factor scores extracted from the factor models. Separate partially latent structural regression models were used to examine associations between experience and frequency (as single exogenous indicators) with DA and MA.

Models were estimated using maximum likelihood with missing values (MLMV) (missing 10% in final models). Model fit was assessed by \({\chi }^{2}\) tests, comparative fit index (CFI), and root mean squared error of approximation (RMSEA). An insignificant \({\chi }^{2},\) CFI > 0.90, and RMSEA < 0.08 indicate a good model fit [28]. Focal strains in the solution were assessed with modification indices. As robustness checks, models were additionally estimated using MLMV with standard errors clustered by clinic; ML with ordinal family and logit link function to account for the Likert-scale of indicators; and mean- and variance-adjusted weighted least squares (WLSMV) with categorical indicators and standard errors clustered by clinic.

The online survey was constructed in Corporater Surveyor, an online survey tool pre-approved by the Data Protection Officer of Haukeland University Hospital. Analysis and data visualization was performed in Stata SE 17, except WLSMV estimated in Mplus v8.1.

Results

Descriptive results

In total, 674 respondents representing 77 (88%) of the 88 outpatient clinics in the Norwegian CAMHS completed the survey. Among these, 484 (72%) of respondents were physicians (including child and adolescent psychiatrists) or licensed clinical psychologists, the two professional groups that are authorized to diagnose patients in Norway. The majority (60%) indicated that they have worked in the CAMHS for 5 years of more, and 84% estimated that they see ADHD patients at least weekly (see Table 1). We estimated the response rate among clinicians to be approximately 38%. The 12% of non-responding clinics were geographically spread around the country, representing both urban and rural catchment areas.

As shown in Tables 2, 3 and 4, there was large variation in responses to most attitude items, indicating that the survey successfully measured a spectrum of attitudes from restrictive to liberal. Generally, responses leaned toward the restrictive rather than liberal end of the scale on most survey items. For example, we see from item ideal (Table 2) that 92% of respondents think that ADHD would be less prevalent had all children lived under ideal conditions when growing up, suggesting that most clinicians think environmental factors should be carefully considered before making a diagnosis.

The provided comment boxes were frequently utilized by respondents, indicating a high degree of engagement. Comments often emphasized the complexity of decision-making in relation to ADHD symptoms, or moderated the forced responses given to the items. Spontaneous feedback from heads of clinics during the distribution of the survey supported the impression that rates of ADHD diagnosis and medication is considered a controversial and important topic.

Tables 1, 2, 3 and 4 show (translated) questions and options as they were presented in the survey, except for headings in italics and symbols to indicate scoring added for clarity. Except for item D8, there was less than 5% missing responses to all items.

Two-factor model of diagnosis and medication attitudes

In the final SEM model, attitude toward diagnosis (DA) was estimated by seven sub-items of the item await (D1–D7), describing situations/conditions that might make clinicians hold back in the decision-making process of diagnosing patients with ADHD. Model fit statistics for the one-factor model for DA (MLMV) were considered acceptable (\({\chi }^{2}\)(11) 65.5, p < 0.001, CFI 0.97, RMSEA 0.086). Attitude toward medication (MA) was comprised of seven sub-items consisting of statements about the treatment of ADHD (medication). Fit statistics indicated good model fit \(({\chi }^{2}\)(13) 36.0, p < 0.001, CFI 0.95, RMSEA 0.051). All factor loadings were significant.

Overall, DA and MA seemed to reflect separate latent constructs, that are moderately correlated (r = 0.4). Model fit statistics for the two-factor model were good (\({\chi }^{2}\)(72) 188.8, p < 0.001, CFI 0.96, RMSEA 0.049). Main results were consistent across model specifications.

Figure 1 presents the two-factor CFA model with standardized parameters; Tables 3 and 4 show details about the indicators. Standardized and unstandardized parameter estimates can be found in Appendix B.

Fig. 1
figure 1

Two-factor SEM model of attitude toward diagnosis (DA; Table 3) and medication (MA; Table 4)

Full size image

Group differences

Intraclass correlations (ICC) indicated that respondents’ professional background constituted 5.2% (95% CI 1.4–17.3) of variation in DA and 1.6% (95% CI 0.2–10.1) in MA. For clinic, ICCs showed that 2.7% (95% CI 0.5–13.7) of variation in DA and 9.8% (95% CI 0.49–18.7) in MA was attributable to the clinic level. Figure 2 shows distribution of the DA and MA factor scores across individuals and clinics, illustrating that individual scores vary more widely than scores based on clinic averages. n = 76 in clinic-level analysis as two closely affiliated clinics were merged to one due to a mix-up of survey links.

Fig. 2
figure 2

Kernel density plot showing variation in DA and MA scores by clinicians (a) and by clinic (b)

Full size image

Partially latent structural regression models indicated that longer experience was associated with somewhat more liberal scores on DA (std. coef. = 0.11, p = 0.04), while it did not affect MA scores (std. coef. = 0.03, p = 0.66). There was no support for an association between frequency and DA (std. coef. = −0.06, p = 0.18), while lower frequency of contact with ADHD patients led to slightly more restrictive MA scores (std. coef. = –0.12, p = 0.03).

Discussion

Our results are based on answers from 674 respondents from the entire population of clinicians working with ADHD in the Norwegian CAMHS outpatient clinics, representing 88% of the clinics in the country. Responses indicated that attitudes toward ADHD diagnosis and medication vary considerably among individual clinicians. Generally, respondents tended to lean toward the restrictive end of the continuum, while a strong liberal attitude was seldom expressed. Only small proportions of variation could be ascribed to professional background or clinic level.

Our CFA supported the hypothesis that attitudes toward diagnosis (DA) and medication (MA) of ADHD can be conceptualized as two separate, but related constructs. The exclusion of the items certainty and ideal signify that our final DA model became somewhat more narrowly defined than originally planned, consisting only of the sub-items of await. Interestingly, when responding to the item certainty, many respondents commented that other circumstances must be taken into consideration and mentioned specifically the topics covered in await, the subsequent item in the survey. Several respondents commented that they answered on behalf of their team, and that they do not necessarily personally agree with every decision. This could help explain why the items certainty and await correlated less than expected even if respondents seem to consider them related. The items in the final model were considered an adequate measure of attitude toward diagnosis, indicative of practical diagnostic decision-making. With regard to the MA model, several aspects of attitude toward medication are covered, including beliefs about effect, side effects, and relation to alternative forms of treatment.

Partitioning the total variance into individual and group level revealed that most variance occurred between individuals, and only small proportions (< 10%) of variance could be ascribed to the group level. Clinic affiliation explained considerably more variance in MA than in DA, suggesting that attitude toward medication is more sensitive to the development of local practice cultures than attitude toward diagnosis. In summary, there seems to be large differences in attitudes between individual clinicians, which is likely to be of consequence to their clinical practice. However, as little variance was attributable to the clinic level, the hypothesis that these attitudes influence local practice cultures and cause the geographic variation in ADHD diagnosis and medication rates was not supported by these results.

To our knowledge, this is the first study aiming to estimate and compare clinician attitudes toward ADHD diagnosis and medication by a comprehensive nationwide survey, and that uses statistical analyses that account for potential variation due to clinic affiliation. Our results are in line with the main findings of related previous studies. Using qualitative methods, Kovshoff et al. (2012) found that clinicians describe the process of making decisions in relation to ADHD as challenging, and that they include in their clinical decision a [subjective] consideration of whether the label will be in the best interest of the child [29]. In a follow-up study, the same researchers showed that clinicians differed in their partiality toward medication versus psychosocial intervention, pointing out that this may lead to patients receiving different advice based on provider preference alone [25]. Further, it was observed that clinicians vary in their attitudes regarding the goal of treatment (optimizing versus being satisfied with some “degree of improvement”), which is likely to cause differences in clinical practice. An earlier, related survey on physician perceptions on ADHD medication showed that physicians generally consider ADHD medication to be effective [23]. However, like us, they also found that clinicians were divided in their assessments of the severity and manageability of side effects.

While relevant, the abovementioned studies all include physicians only, albeit of different specialties. To the extent that we can compare our results to studies that include a respondent group including clinicians of diverse educational backgrounds, there seems not to be consistent patterns of group differences in attitudes related to the background variables we have collected. An ambitious international survey [27], although hampered by very low response rate, showed differences between professions in their concerns related to rates of ADHD diagnosis: psychologists were most worried about overdiagnosis, and psychiatrists about underdiagnosis. Dekkers et al. (2021), on the other hand, considered factors related to clinicians’ policy and attitudes toward medication and parent training, and found no difference between clinicians of medical versus non-medical backgrounds in attitudes toward medication or how likely they were to discuss parent training or medication as potential treatment options [26]. This is in line with our finding that professional background explains less than 2% of the observed variation in medication attitude.

Strengths and limitations

The survey used in this study was designed specifically to extract variation in our hypothesized “restrictive to liberal” attitude dimension. The results, showing large differences between individuals, suggest that we succeeded in this regard. To maximize validity and reliability, the survey was constructed with the involvement of several experienced clinicians, both internal and external to the research group, and was piloted before distribution. The main strength of the study is that we were able to reach out to the entire population of clinicians, and that we achieved a high number of respondents, a satisfactory response rate, and very high geographical, clinic-wise coverage. We have also gathered background variables enabling us to investigate group differences in responses between several relevant groups.

Regarding limitations, we do not know if there are shared characteristics within the group of respondents or non-respondents. It is possible that those heads of clinics and clinicians who chose to invest time in the survey represent those particularly concerned about overdiagnosis and overtreatment of ADHD. Interestingly, despite being of high professional relevance, the topic of this survey seems to be considered sensitive by many. This can potentially have lowered the response rate or skewed answers in a direction considered more socially acceptable by respondents.

Another possible threat to the validity of any survey pertains to the wording and order of questions and options, as this will shape the collected information. As a specific example, the abstract wording of the options of the item certainty seemed to anchor respondents to the first, most restrictive alternatives. If we had reworded to emphasize that 50% certainty corresponds to “finding it more likely that ADHD is the correct diagnosis than not”, the response distribution would likely have been different.

Lastly, we cannot know for certain how much of the actual variation we have succeeded in extracting with our items, and we do not know how respondents would truly act in practical situations. It may be that clinicians who consider themselves similarly restrictive or liberal according to our scale will still make different decisions when faced with comparable cases in clinical practice.

Together, these challenges may explain why our findings did not show clear associations between clinicians’ individual attitudes and group affiliation, leaving the question about clinical practice cultures as a reason for geographical variation in rates of ADHD still unsettled.

Conclusion

This survey has succeeded in measuring attitudes toward ADHD diagnosis and medication among Norwegian clinicians. We have demonstrated large variation in attitudes between clinicians working in the field, which in turn suggests that clinical practice varies between individual clinicians. However, we did not find large group differences based on clinic affiliation or other background variables. Our results thus do not point to variation in individual attitudes as a main reason for the geographic variation in diagnosis and medication rates. While this alone does not rule out local practice cultures as a reason for the variation, it suggests that any existing practice cultures are driven by forces other than conscious or idealistic differences that we have studied in this survey. As this is an explorative study, its results must be interpreted with some caution. There is need for further research on the potential influence of practice culture on the observed geographic variations in prevalence and treatment of ADHD.