Electroconvulsive therapy following incident bipolar disorder: When, how, and for whom?

Abstract

Objective

The use of electroconvulsive therapy (ECT) in the treatment of bipolar disorder (BD) remains poorly described. Based on data from Danish registries with complete nationwide coverage, this study of patients with incident BD aimed to describe when, how, and for whom ECT is used in the context of BD.

Methods

We identified patients receiving their first diagnosis of BD in the period from 2008 to 2018, who subsequently received ECT. Descriptive statistics were used to clarify when, how, and for whom ECT is used.

Results

We identified 1338 patients with incident BD who subsequently received ECT. The median age at the first ECT session was 50.6 years (interquartile range [IQR]: 26.4), and 62% of those treated with ECT were female. The median time from the diagnosis of BD to the first ECT treatment was 0.6 years (IQR: 2.6), and 58% of the patients receiving ECT had the first treatment within the first year after being diagnosed with BD. The most common indication for the first ECT treatment was depression (mainly non‐psychotic depression), followed by mania (mainly psychotic mania). The first ECT session was typically provided to inpatients (97%), upon patient consent (98%) and with bilateral electrode placement (60%).

Conclusions

A substantial proportion of the patients with incident BD who receive ECT require this treatment within the first year after the diagnosis. The most common indication for ECT is depression followed by (psychotic) mania. Inpatient voluntary ECT using bilateral electrode placement is the most common form of administration.

1. INTRODUCTION

While pharmacotherapy represents the backbone of the acute and maintenance treatment of bipolar disorder (BD), ¹ electroconvulsive therapy (ECT) is usually recommended as second‐line treatment for patients with treatment‐resistant conditions or when pharmacotherapy cannot be tolerated. ¹ ECT can also be used as first‐line treatment for acutely severe conditions in the context of BD, such as catatonia, aggressive behavior, imminent suicidality, or in cases where pharmacotherapy might be undesirable (e.g., pregnancy). ¹ , ²

Despite the fact that ECT is usually considered as an alternative to pharmacological treatment in BD, there is solid evidence for its effectiveness in the treatment of manic, depressive and mixed episodes. ³ , ⁴ , ⁵ , ⁶ For instance, ECT has consistently shown a rapid effect in patients with severe treatment‐resistant mania, especially in cases of delirious mania or excited catatonia. ⁴ , ⁷ Nevertheless, a comprehensive examination of the patterns of use of ECT in BD in real‐world settings is lacking from the literature. Moreover, since ECT is less commonly used in the early course of the illness, ¹ the use of ECT in patients recently diagnosed with BD remains poorly described.

In the present study, we aimed to describe the use of ECT in a nationwide cohort of patients with incident BD using data from the Danish registries. Specifically, we aimed to address the following research questions: (i) when are patients first referred to ECT after the diagnosis of BD? (ii) how is ECT usually administered? and (iii) who starts ECT in the first years after the diagnosis of BD?

2. MATERIALS AND METHODS

2.1. Design

We performed a population‐based cohort study using nationwide data from Danish registries. ⁸ Compared to other European and non‐European countries, Denmark has a relatively high ECT utilization, ⁹ , ¹⁰ and the administration of ECT is quite standardized across psychiatric services in accordance with the national clinical guidelines for ECT published by the Danish Psychiatric Society. ¹¹

2.2. Registers

This study was based on data from two nationwide Danish registries: the Danish Civil Registration System (DCRS) ¹² and the Danish National Patient Registry (DNPR). ¹³ The DCRS was established in 1968 and contains information about the date of birth, birthplace, and vital status of all individuals with legal residence in Denmark. ¹² It also includes a unique 10‐digit personal identification number, which enables linkage of information (at the level of the individual) across all national Danish registries. The DNPR contains International Classification of Disease, 10th Revision (ICD‐10) diagnostic codes from all admissions to Danish non‐psychiatric and psychiatric hospital services since 1994 as well as from outpatient contacts and emergency room visits since 1995. ¹³ Registration of all ECT sessions has been mandatory since 2001. However, data on ECT treatments in the DNPR are assumed to be stable and credible only from 2008 and onwards, according to a report from the Danish National Board of Health. ¹⁴ As there are no private hospitals or clinics in Denmark offering ECT, ⁹ the information in the DNPR covers all treatments given in the country.

2.3. Study population

In order to obtain a population with incident BD, we used the DNPR to identify all patients receiving their first diagnosis of BD (ICD‐10: F30–F31) in the period from January 1, 2008, to December 31, 2018. From this group of patients, we identified those who received ECT—after the diagnosis of BD—in the period from January 1, 2008, to December 31, 2018. Patients receiving their first diagnosis of BD in the period from January 1, 2008, to December 31, 2018, but not receiving ECT after the diagnosis were included as reference group. For each patient, we obtained information on the date of birth, sex, age at diagnosis of BD, and any psychiatric and non‐psychiatric hospital admissions and psychiatric diagnoses in the 2 years preceding the date of the diagnosis of BD. Finally, we defined presence of a chronic medical disease in the 2 years preceding the date of the diagnosis of BD in accordance with the Charlson Comorbidity Index. ¹⁵

2.4. Definition of when, how and for whom

In order to describe when, how, and for whom ECT is being used in the context of incident BD, the following definitions were made:

When—the time from the diagnosis of BD to the first ECT session: For those receiving ECT, we obtained information on the time elapsed from the diagnosis of BD to the first ECT session, as well as on the age at which they received this treatment. To minimize right truncation due to the end of follow‐up on December 31, 2018, the estimation of time to ECT and age at ECT was based only on the subset of patients receiving the diagnosis of BD from January 1, 2008, to December 21, 2014. We also provided a characteristic of those receiving ECT in the first year following BD.

How—the administration of the first ECT treatment: For this purpose, we extracted information on the following characteristics of the first ECT session, that is, setting (outpatient or inpatient), electrode placement (unilateral [procedure code: BRTB10, BRXA10], bilateral [procedure code: BRTB11, BRXA11] or unspecified [procedure code: BRTB1, BRXA1]), and consent (voluntary [procedure code: BRXA*] or involuntary [procedure code: BRTB*]). ¹⁶

For whom—the indication for ECT: We retrieved information on the indication diagnosis for the first course of ECT and grouped them as follows: hypomania (ICD‐10: F30.0, F31.0), mania without psychotic symptoms (ICD‐10: F30.1, F31.1, F30.8, F30.9), mania with psychotic symptoms (ICD‐10: F30.2; F31.2), depression without psychotic symptoms (ICD‐10: F31.3, F31.4, F32.0–F32.2, F32.8, F32.9, F33.0–F33.2, F33.4, F33.8, F33.9), depression with psychotic symptoms (ICD‐10: F31.5, F32.3, F33.3), mixed episode (ICD‐10: F31.6), other bipolar episodes (ICD‐10: F31.7, F31.8, F31.9), psychotic disorders (ICD‐10: F20–F29), organic mental disorder (ICD‐10: F00–F09), and any other indication.

The indication diagnoses for any acute courses of ECT following the first ECT course were also examined (for a definition of an acute course of ECT, please see the Supplementary Material; Data S1). For this analysis, we only included the subgroup of patients initiating ECT between January 1, 2008, and December 31, 2014, to ensure that they all had four complete years of follow‐up, and therefore, all could—potentially—contribute to the full follow‐up period.

2.5. Statistical analyses

Descriptive statistics were used to characterize the study population after removing duplicates. ⁹ Since the aim of the study was to gain more insight on the use of ECT for BD in a real‐world setting, without seeking to make inferences, statistical comparisons between groups were not deemed relevant. The group of patients with incident BD not receiving ECT was only included as a reference. All analyses were performed in STATA (version 16.1).

2.6. Ethics

The use of the register‐based data for the purpose of this study was approved by Statistics Denmark and the Danish Health Data Authority. No further ethical approval is required for register‐based research in Denmark. The project is registered on the internal list of research projects having the Central Denmark Region as data responsible.

3. RESULTS

3.1. Study population

We identified a total of 20,295 patients registered with their first (incident) ICD‐10 diagnosis of BD in the period from January 1, 2008, to December 31, 2018, of whom 1338 (6.6%) received ECT in the period from January 1, 2008 to December 31, 2018 (median number of sessions in the first course of ECT: 8, interquartile range [IQR]: 6). The characteristics of the study population are shown in Table 1. Compared to the patients with incident BD, who did not receive ECT (median age of 39.4 years, IQR: 25.2), those receiving ECT (median age of 48.5 years, IQR: 27.3) were substantially older at the time of the diagnosis of BD, especially those receiving ECT in the first year following the diagnosis of BD (median age of 51.3 years, IQR: 27.5). Notably, a substantial fraction (29.5%) of the patients with BD receiving ECT had previously been diagnosed with unipolar depression in psychiatric (hospital) services.

TABLE 1.

Characteristics of the study populations.

	Bipolar disorder without subsequent ECT (n = 18,957)	Bipolar disorder with subsequent ECT (n = 1338)	ECT in the first year following bipolar disorder (n = 840)
Age at diagnosis, years, median (IQR)	39.4 (25.2)	48.5 (27.3)	51.3 (27.5)
Sex, female (%)	10,954 (57.8)	828 (61.9)	512 (61.0)
Age at first ECT, years, median (IQR)	‐	50.6 (26.4) a	51.3(27.3)
Time from diagnosis to first ECT, years, median (IQR)	‐	0.6 (2.6) a	‐
Prior admission, n (%) b
Psychiatric admission	4433 (23.4)	469 (35.0)	294 (35.0)
Non‐psychiatric admission	6369 (33.6)	515 (38.5)	335 (39.9)
Previous psychiatric diagnoses, n (%) b
Organic mental disorder	303 (1.6)	30 (2.2)	19 (2.3)
Psychotic disorder	1319 (7.0)	128 (9.6)	69 (8.2)
Unipolar depression	3369 (17.8)	395 (29.5)	275 (32.7)
Anxiety disorder	2831 (14.9)	225 (16.8)	142 (16.9)
Eating disorder	120 (0.6)	11 (0.8)	X
Personality disorder	839 (4.4)	59 (4.4)	29 (3.4)
ADHD	350 (1.8)	16 (1.2)	8 (0.9)
Charlson comorbidity index diseases, n (%) b
Charlson 0	16,919 (89.2)	1173 (87.7)	723 (86.1)
Charlson 1	1507 (7.9)	127 (9.5)	88 (10.5)
Charlson ≥2	534 (2.8)	38 (2.8)	29 (3.4)

Note: The primary psychiatric diagnoses were coded based on ICD‐10 in the following categories: organic mental disorder (F00–F09); psychotic disorder (F20–F29); unipolar depression (F32–F33); anxiety disorder (F40–F48); eating disorder (F50); personality disorder (F60–F69); and ADHD (F90). X: cannot be listed due to small numbers and risk of identification of individuals.

Abbreviations: ADHD, attention‐deficit/hyperactivity disorder; ECT, electroconvulsive therapy; IQR, interquartile range.

^a Only patients diagnosed with bipolar disorder between 2008 and 2014 (n = 1026) were included in this calculation to ensure that all patients had at least 1 year of follow‐up (limited right truncation).

^b Considering the 2 years preceding the diagnosis of bipolar disorder.

Figure 1A shows, for each year from 2008 to 2017, the proportion of patients receiving ECT within the first year after the diagnosis of BD from all the patients receiving a diagnosis of BD the same year. Both the proportion and the absolute number of patients referred to ECT in the first year after being diagnosed with BD decreased slightly over the period from 2008 to 2017, but remained stable around 3% from 2014 and onwards. Figure 1B displays the number of patients diagnosed with BD per year in the period from 2008 to 2018, together with the median age at diagnosis for both the whole population of patients with incident BD, the group of patients receiving ECT, and the subgroup referred to ECT in the first year after the diagnosis. The number of patients with incident BD tended to increase throughout the study period, while the median age at diagnosis tended to decrease. The median age at diagnosis for the patients receiving ECT remained around 48 years over the period from 2008 to 2017, while, for the subgroup receiving ECT in the first year after the diagnosis of BD, it remained around 51 years.

FIGURE 1.

Use of electroconvulsive therapy (ECT) in the year following the diagnosis of bipolar disorder throughout the study period. (A) The number and proportion of patients with incident bipolar disorder receiving ECT in the first year following the diagnosis. (B) The number of patients with incident bipolar disorder throughout the study period (bars) and the median age at diagnosis for this population as well as for the patients receiving ECT (grey and blue lines).

3.2. When, how, and for whom

3.2.1. When—the time from the diagnosis of BD to the first ECT session

The median time from the date of the diagnosis of BD to the first ECT session was 0.6 years (IQR: 2.6) (Table 1).

3.2.2. How—the administration of the first ECT treatment

The vast majority of patients received their first ECT treatment as inpatients (97.3% inpatients vs. 2.7% outpatients) and only 2.2% were treated involuntarily. Regarding the electrode placement for the first ECT session, 60.4% of the patients received bilateral ECT, 15.4% of the patients received unilateral ECT, and for 24.2% of the patients, the electrode placement was not specified. Involuntary ECT was predominantly bilateral (69.0% of the first treatments, while the remaining 31.0% were unspecified).

3.2.3. For whom—the indication for ECT

Figure 2 shows the indication diagnoses for the first course of ECT for the patients with incident BD. The most common indication was depression (41.6% non‐psychotic depression and 14.8% psychotic depression), followed by mania (4.6% non‐psychotic mania and 7.5% psychotic mania), other bipolar episodes (9.8%) and mixed episodes (9.3%). When comparing indication diagnoses for ECT between the sexes, the most common indication for the first ECT session was depression for both females (54.1%) and males (59.9%), but a larger proportion of females than males received the first ECT due to a mixed episode (11.0% vs. 6.5%) (Figure 2A). The indication diagnosis for ECT stratified by age groups is shown in Figure 2B. The proportion of patients having depression as the indication for ECT tended to increase with age (42.2% among those aged ≤30 years vs. 64.4% among those over the age of 60), while the proportion of patients with psychotic mania or a mixed episode tended to decrease (28.3% among those aged ≤30 years vs. 9.4% among those over the age of 60).

FIGURE 2.

Indication diagnoses for the first course of electroconvulsive therapy (ECT) stratified by sex and by age. (A) Indication diagnoses for the first course of ECT stratified by sex. (B) Indication diagnoses for the first course of ECT stratified by age at first ECT. The number of patients per age group is as follows: ≤30 years = 223; 31–40 years = 208; 41–50 years = 267; 51–60 years old = 247; ≥61 years old = 393. “Organic mental disorder” is included under the label “Other disorders.”

From the subset of patients with incident BD receiving ECT for the first time in the period from 2008 to 2014 (n = 774), a total of 236 patients (30.5%) received one or more acute ECT courses during follow‐up. Of those, 39.1% were delivered during the first year following the first course of ECT. Figure 3 shows the distribution of indication diagnoses for ECT in the 4 years after the first ECT (included as “year 0” for comparison). Non‐psychotic depression remained the main indication for ECT, while psychotic depression and mania were less common indications for ECT during follow‐up. Conversely, “Other disorders” became a more common indication diagnosis for ECT in the years following the first ECT. Notably, this mixed category includes psychotic disorders, which accounted for 6.3% of the diagnostic indications for acute ECT courses in the year after the first ECT, 20.7% in the second year, and 14.4% in the third and fourth year (merged to avoid risk of identification of individuals).

FIGURE 3.

Indication diagnoses for the first course of electroconvulsive therapy (ECT) and the subsequent courses of acute ECT received during 4 years of follow‐up. Year 0 represents the indication diagnoses for the first course of ECT for the individuals starting ECT between 2008 and 2014 (n = 774). Year 1 includes the indication diagnoses for all acute ECT courses in the year following the first course of ECT (n = 207). Years 2, 3 and 4 include the indication diagnoses for all acute ECT courses during the second (n = 82), third (n = 45) and fourth (n = 52) year after the first ECT, respectively. Further (specific) diagnostic categories cannot be listed due to small numbers and risk of identification of individuals. As a consequence, “Psychotic mania,” “Non‐psychotic mania” and “Mixed episodes” have been merged under the label “Manic/mixed episodes.” “Other bipolar episodes,”, “Psychotic disorders,” and “Organic disorders” are included under the label “Other disorders.”

4. DISCUSSION

This is the first study to explore the use of ECT in a nationwide cohort of patients with incident BD. The main findings of the study are as follows: (i) a substantial proportion of the patients with newly diagnosed BD who received ECT required this treatment in the first year after the diagnosis; (ii) patients receiving ECT were relatively old at the time of the diagnosis of BD, and many had been diagnosed with unipolar depression in psychiatric (hospital) services prior to being diagnosed with BD; (iii) the majority of the patients with BD received ECT as inpatients, voluntarily, and with bilateral electrode placement; (iv) depression was the far most common indication for the first course of ECT among patients with BD; (v) mixed episode was a relatively common indication for starting ECT among women and younger patients; and (vi) depression remained the main indication for ECT treatment beyond the first course of ECT, while an increasing—although modest—proportion had psychotic disorder as indication for ECT.

4.1. When—the time from the diagnosis of BD to the first ECT session

Our results showed that more than 60% of the patients with incident BD required this treatment within the first year after the diagnosis. Those referred to ECT in the year after the diagnosis were older and almost one third of them had been diagnosed with depression in the 2 years preceding the diagnosis of BD. These characteristics suggest that these patients are likely to be more chronically ill and resistant to pharmacological treatment, which may explain the choice of ECT in this subset of patients. Furthermore, these characteristics could suggest that a substantial fraction of this subgroup may have had undetected BD (mistaken for unipolar depression), which has likely led to a delay in adequate pharmacological treatment with a mood stabilizer, ¹⁷ that is, a relatively long duration of untreated illness. ¹⁸ As there is a positive association between the duration of untreated illness in BD and poor clinical outcome, including higher number of manic and depressive recurrences and higher suicidal risk, ¹⁸ , ¹⁹ long duration of untreated illness may explain why those requiring ECT within the first year after the diagnosis of BD were relatively old at the time of diagnosis. This underscores the importance of screening for prior (hypo)manic/mixed episodes in patients with depression. While a large proportion of these patients will not have experienced any symptoms suggestive of BD, for the minority in which such a screening turns out positive, it will likely reduce the duration of untreated BD, thereby improving the prognosis. ¹⁸ , ¹⁹ , ²⁰ According to our data, the number and proportion of individuals requiring ECT in the year after the diagnosis of BD has, however, experienced a slight decline over time. This could represent an effect of earlier diagnosis of BD, as we have shown that the age at diagnosis of BD has decreased over the study period, which is in line with findings from other studies of BD also based on data from the Danish registries. ¹⁷ , ²¹ This means that patients are probably less likely to be chronically ill/treatment resistant at the time of diagnosis, which, in turn, may reduce the need for ECT. Relatedly, it may also be a consequence of an increasing proportion of patients being diagnosed with BD type II, a subgroup who typically experiences less severe acute episodes requiring ECT. ¹⁷ The latter possibility can, however, not be explored based on the data at hand, as the distinction between BD type I and II is not made in the ICD‐10. Notably, this same tendency was not observed in the subset of patients with incident BD who received ECT within the first year after the diagnosis, for whom the median age at diagnosis remained fairly stable around 51 years.

4.2. How—the administration of the first ECT treatment

Regarding the administration of ECT in incident BD, our results are similar to the findings of previous register‐based studies examining the use of ECT across a wider range of mental disorders. ⁹ The most common electrode placement was bilateral, and most patients received their first ECT during an admission, likely reflecting the severity of the condition leading to the referral to ECT. Hence, once the decision of prescribing ECT has been made by mutual consent with the patient (whenever feasible), clinicians seem to prioritize efficacy over tolerability, that is, bilateral over unilateral electrode placement. Also, we found that involuntary ECT was rarely used for patients with incident BD. In Denmark, involuntary ECT can only be used when a patient is at imminent risk of dying due to severe states of, for example, dehydration following refusal/inability to eat/drink, malignant catatonia or delirium, for which ECT can be life‐saving. ¹⁶ , ²²

4.3. For whom—indication for ECT

Depression was the far most common indication for ECT in this population of patients with incident BD, both for the first course of ECT and for subsequent courses of ECT. This finding could be explained by the natural course of BD, with a predominance of depressive symptoms, ²³ but it could also reflect the differences in the number of therapeutic options available for the opposing mood poles of BD. Specifically, there are several pharmacological agents approved for the treatment of mania, and up to 80% of patients with mania respond to either monotherapy or a combination of these medications, which is why ECT is typically recommended in case of lack of response to pharmacological treatment or in very severe cases of acute mania. ⁷ Accordingly, we observed that ECT was more commonly used to treat psychotic mania compared to non‐psychotic mania. Conversely, treating bipolar depression is usually more problematic because of lower response rates to pharmacological treatment and a more limited number of evidence‐based first‐line options, ²⁴ , ²⁵ with only three pharmacological treatments (lithium, lamotrigine, and quetiapine) approved for bipolar depression by the Food and Drug Administration in the period under study. ²⁵ This may lead to ECT being considered earlier in the treatment (hierarchy) of bipolar depression compared to mania. From an international comparative perspective, our results indicate that the use of ECT for manic episodes in Denmark is quite high compared to other countries. ¹⁰ Despite these differences in the use of ECT in mania and depression, its efficacy in both mood states is well established. ⁴ Relatedly, while we did not include outcome indicators for ECT in the current study, a prior mirror‐image study from our group, also based on data from the Danish national registries, showed that ECT indeed seems to have a beneficial effect on self‐harm/suicide attempts in the real‐world clinical setting (as in clinical studies) in BD, as well as for other diagnostic groups. ²⁶

As reported by other register‐based ²⁷ , ²⁸ and clinical ²⁹ , ³⁰ studies, our findings support that ECT is more commonly used among females than among males in the context of BD. While depression was the most frequent indication diagnosis for the first course of ECT in both sexes, the proportion of males receiving ECT due to depression was slightly higher compared to females, while the proportion of males and females receiving ECT due to mania was very similar between the sexes. This finding was somewhat unexpected since women with BD tend to report more depressive symptoms than men. ³¹ , ³² One explanation for this finding, as pointed out by Karanti et al., ²⁹ could be that females are more likely to suffer from subsyndromal depressed mood, that is not an indication for ECT, but there might be no sex‐differences in the number of full‐blown depressive episodes or in their severity. ³² Notably, we found that, compared to their male counterparts, a larger proportion of female patients with incident BD received ECT due to a mixed episode, which is consistent with the literature describing that mixed episodes are more common in females. ³³ Mixed episode was also a common indication for first ECT among younger patients, together with psychotic mania, while depression and non‐psychotic mania became more common indications with increasing age. This is in agreement with cohort studies reporting more psychotic features, ³⁴ , ³⁵ , ³⁶ higher risk of hetero‐aggressive behavior, ³⁴ and more mixed episodes ³⁷ among patients with early‐onset BD compared to those with a later onset. Also, some studies describe depressive episodes to become more frequent in the course of BD with increasing age, ³⁸ while manic symptoms tend to be milder in older patients with BD. ³⁹ Taken together, these results likely reflect that, in younger patients, ECT is commonly prescribed due to acute and severe conditions with prominent behavioral alterations, while treatment resistance might be a more common indication in older patients.

The choice of ECT for the treatment of depression and mania among older patients recently diagnosed with BD might be also linked to the presence of medical comorbidities, which may limit pharmacological treatment options, or to the fact that elderly patients may show a poorer and/or slower response to pharmacological agents. ⁴⁰ , ⁴¹ Also, there is increased risk of suicide in later life. ⁴² In all these cases, use of ECT may be indicated. Data on the use of ECT in old age BD are, however, quite scarce, and there may be some concerns for potential neurocognitive side effects with ECT in this more vulnerable population, which, in some countries/settings, can lead clinicians to refrain from referring elderly patients to ECT. Without being direct evidence of its efficacy and safety, our documentation of the fact that ECT is commonly used for the treatment of bipolar depression and mania in older patients with BD in Denmark indirectly suggests that the benefit of ECT for these patients outweighs the risks and side effects. This is in agreement with the available literature on this topic. ⁴³ , ⁴⁴ , ⁴⁵

4.4. Strengths and limitations

The main strength of this study stems from its register‐based approach that allowed us to identify more than 1300 patients with incident BD who subsequently received ECT. There are, however, also a number of inherent limitations associated with the use of these register‐based data. ⁴⁶ First, the DNPR only contains indication diagnoses for ECT, and we were therefore not able to investigate the more specific clinical rationale for choosing ECT (e.g., treatment‐resistance, suicidal ideation, catatonia, or self‐ or hetero‐aggressive behavior). Second, data on symptoms and cognition are not available in the Danish national registries, and these outcomes could therefore not be explored. Third, the patients in this study have not been systematically assessed using standardized diagnostic tools (in the context of this study). Several diagnoses from the registers—including unipolar depression, BD, and schizophrenia—have, however, been validated with satisfactory results. ⁴⁷ , ⁴⁸ , ⁴⁹ Fourth, to avoid potential identification of individuals, it was necessary to group patients under the age of 30 years and those above the age of 60 years, respectively, yielding less fine‐grained results for the young and the elderly. Fifth, as the distinction between bipolar I and bipolar II is not possible in ICD‐10, analyses stratified by these subtypes could not be carried out. Finally, and most importantly perhaps, this study was based solely on data from Denmark, and the results will therefore not generalize to other countries/populations in some aspects. ¹⁰

In summary, this nationwide study of the use of ECT in incident BD indicates that ECT is often used within the first year after the diagnosis and that diagnostic indications differ according to sex and age at diagnosis of BD. The most common indication for ECT is depression followed by mania, while inpatient voluntary ECT using bilateral electrode placement is the most common form of administration. Furthermore, as (i) the patients who require ECT within the first year after being diagnosed with BD are relatively old at the time of the diagnosis and (ii) a substantial proportion of them have a recent history of hospital‐treated unipolar depression, there may be a need for more careful screening for bipolarity in patients with unipolar depression – as this may in fact be bipolar depression.

CONFLICT OF INTEREST

Dr Rohde received the 2020 Lundbeck Foundation Young Investigator Prize. Dr Østergaard received the 2020 Lundbeck Foundation Young Investigator Prize. Furthermore, Dr Østergaard owns units of mutual funds with stock tickers DKIGI and WEKAFKI, as well as units of exchange‐traded funds with stock tickers TRET and EUNL. Dr Vieta has received grants and served as consultant, advisor or CME speaker for the following entities: AB‐Biotics, AbbVie, Angelini, Biogen, Boehringer‐Ingelheim, Celon Pharma, Dainippon Sumitomo Pharma, Ferrer, Gedeon Richter, GH Research, Glaxo‐Smith Kline, Janssen, Lundbeck, Novartis, Orion Corporation, Organon, Otsuka, Sage, Sanofi‐Aventis, Sunovion, and Takeda, outside the submitted work. Dr Salagre reports no financial relationships with commercial interests.

Supporting information

Data S1

Salagre E, Rohde C, Vieta E, Østergaard SD. Electroconvulsive therapy following incident bipolar disorder: When, how, and for whom? Bipolar Disord. 2022;24:817‐825. doi: 10.1111/bdi.13254

DATA AVAILABILITY STATEMENT

The data used for this study cannot be shared according to Danish Law. Access to the data can be achieved for researchers affiliated with Danish institutions upon application to Statistics Denmark and the Danish Health Data Authority.