Chronic mental health problems and use of mental health services among adults with and without adverse childhood experiences: a 6-year longitudinal population-based study

Abstract

Background

The negative effects of single and multiple adverse childhood experiences (ACE) on adult mental health are well-documented. However, little is known about the prevalence of chronic mental health problems (MHPs) and use of mental health services (MHSs) compared to adults without an ACE history.

Aims

Examine differences in the prevalence of chronic MHP and MHS use between adults without and with a single and multiple ACE history, and MHS use among ACE and no-ACE adults with chronic MHP.

Method

A 6-year longitudinal study was conducted based on a random sample of the Dutch adult population (n = 2427).

Results

Multivariate logistic regression analyses showed that adults with a single (n = 280) and multiple ACE (n = 92) history suffered more often from chronic MHP such as chronic depressive and anxiety symptoms. Both ACE subgroups also made greater use of MHS than no-ACE adults (n = 2055). Multiple ACE compared to single ACE, increased the risk of MHP. However, among those with chronic MHP respondents with and without an ACE history did not differ in MHS use.

Conclusions

ACE history is strongly related to chronic MHP and persistent use of MHS. However, MHS use among those with chronic MHP is not related to an ACE history.

Keywords:

• Adverse childhood experiences
• anxiety and depressive symptoms
• mental health services use
• sleep problems
• fatigue
• longitudinal
• population

Introduction

Adverse childhood experiences (ACEs), such as physical, emotional, and sexual abuse (maltreatment), violence, and death of a significant other may negatively impact child development. Many studies have examined the effects of (recalled) ACE on the mental health during adulthood. Systematic reviews and meta-analyses consistently demonstrated that adults with a (recalled) ACE history are more at risk of mental health problems (MHPs) varying from depression to sleep problems (Hughes et al., 2017; Ip et al., 2016; Kajeepeta et al., 2015; Kalmakis & Chandler, 2015; Liu, 2017; Mandelli et al., 2015; Selous et al., 2020; Zatti et al., 2017).

This consistent finding raises the question to what extent adults with a recalled ACE history more often suffer from chronic MHP compared to adults without an ACE history. According to the stress sensitization model, e.g. that ACE may lead to a lower tolerance for stress during adulthood due to for instance maladaptive coping mechanisms, we might expect that adults with an ACE history also have more difficulties to recover from MHP and consequently more often suffer from chronic MHP (cf. Hammen et al., 2000; Vranceanu et al., 2007). Insight into the chronicity of MHP may therefore further help to disentangle the long-term effects of ACE and determine the prevalence of adults with an ACE history who, because of the chronicity of their problems, may need help the most. The difference between the prevalence of chronic MHP over a long period and the prevalence of MHP at certain moments, referred to as point prevalence, is also of interest. It provides further insight into what extent MHP fluctuates over time (i.e. are absent or present at one or more moments during a given period). In cases where the point prevalence of MHP is considerably lower than the prevalence of chronic MHP, it is relevant to gain an insight into the risk or protective factors that distinguish those with temporary or recurrent MHP from those with chronic MHP. Current ACE studies on risk factors for MHP during adulthood hardly make this distinction.

However, the number of studies assessing chronic MHP as opposed to studies on the prevalence of MHP is limited. ACE studies with regard to chronic MHP have mainly focused on mental disorders, especially depression. For instance, Hovens et al. (2012) showed that emotional neglect and psychological abuse were associated with a chronic course of depression or anxiety disorder. Rhebergen et al. (2011) found that an ACE history was associated with having a depressive and/or anxiety disorder at follow-up 7–9 years after baseline while controlling for physical illness. Fuller-Thomson et al. (2014) found that ACE was predictive of lower depression remission rates compared to adults with depression without an ACE history. Very few ACE studies prospectively examined the prevalence of chronic MHP using random samples of the general population or subpopulations (cf. Brown et al., 1994; McLaughlin et al., 2010; Suija et al., 2011). One such example is the community-based prospective study by Collishaw et al. (2007) with a baseline and a 30-year follow-up assessment, showing that a higher percentage of subjects with a recalled history of abuse suffered from (recalled) recurrent major depressive disorder (i.e. three or more episodes in the past) than those without an abuse history (20.3 versus 3.3%).

To gain more insight into the effects of ACE on chronic MHP, further longitudinal and population-based studies are needed focusing not only on chronic anxiety and depression symptomatology but also on other relevant MHP such as fatigue (cf. de Venter et al., 2017; Heim et al., 2006) and sleep problems (cf. Kajeepeta et al., 2015; Poon & Knight, 2011). Moreover, in light of the fact that physical health problems are positively associated with MHP and that ACE is associated with physical health problems, studies on chronic MHP that take the effects of physical health problems into account are warranted (cf. Rhebergen et al., 2011; Tomasdottir et al., 2015).

Chronic MHPs may also affect the use of mental health services (MHSs). Recent studies on mental health utilization showed that ACE is associated with a higher utilization of MHSs (Bellis et al., 2017; Chartier et al., 2010; Larkin et al., 2018; Mills et al., 2012). However, the extent to which adults with an ACE history and chronic problems are more or less likely to make long-term or “chronic” use of MHS compared to non-ACE adults is unknown. We might expect that, as an outcome of maladaptive coping mechanisms, ACE adults with chronic MHP less often use MHP than non-ACE adults with similar problems.

This brings us to the following research questions:

1. To what extent are adults with an ACE history at greater risk of developing MHPs and chronic MHP than no-ACE adults?

2. To what extent do adults with an ACE history make more use, and more “chronic” use, of MHSs than no-ACE adults?

3. Among adults with chronic MHP, to what extent do those with an ACE history make more frequent use of MHSs than no-ACE adults?

Because the number of different ACE is associated with a higher risk of MHP (Hughes et al., 2017; Selous et al., 2020), ; in this study we made a distinction between those confronted with a single ACE (one type of ACE) and those confronted with multiple ACE (two or more types of ACE).

For this purpose, we conducted a multi-wave population-based study among adults covering a 6-year period. To rule out as much as possible that associations were confounded by other factors that are associated with MHP, such as sex and age (cf. Salk et al., 2017), level of education (cf. Lorant et al., 2003), employment (cf. Kim & von Dem Knesebeck, 2016), marital status (cf. Blanner Kristiansen et al., 2019), and physical illness (cf. Thom et al., 2019) we controlled for these factors.

Method

Procedures and methods

Data were extracted from the Longitudinal Internet studies for the Social Sciences (LISS) panel. The LISS panel is administered by CentERdata (Tilburg University, the Netherlands), funded by the Dutch Research Council (NWO), and based on a large random sample drawn from the Dutch population register by Statistics Netherlands (Scherpenzeel & Das, 2011). Panel members who do not have a computer and/or internet access are provided with the necessary equipment at home. All members receive an incentive of 15 euros per hour for their participation. Further information about the panel and free access to the de-identified data can be found on https://www.lissdata.nl/. In accordance with the new General Data Protection Regulation, participants gave explicit consent for the use of the collected data for scientific and policy relevant research. Proposals for studies such as the first survey were evaluated and approved by the Board of Overseers, the Internal Review Board (2010 proposal nr. 75, Tonic immobility in response to trauma: Prevalence and consequences).

We used data collected in five surveys. For each survey, all panel members were invited to participate. In July 2011 (T1) ACEs were assessed (n^participants = 4613, response = 67.9%); in November–December 2011 (T2) MHP, MHS use and physical problems (n^participants = 5044, response = 77.2%); in November–December 2013 (T3) MHP and MHS use (n^participants = 5343, response = 85.9%); in July–August 2015 (T4) MHP and MHS use (n^participants = 5975, response = 83.8%); and in November–December 2017 (T5) MHP and MHS use (n^participants = 5927, response = 79.2%). In total, 2427 respondents participated in all five surveys.

Non-response analyses

Multivariate logistic regression analyses was conducted to analyze the extent to which ACE (T1), and demographics, MHP, MHS and medicine use, and illness (all at T2) predicted non-response (1= participated at T1 and T2 only, 2 = participated at T1 to T5, N^{T1-T2 total} = 4126). Non-response was not significantly associated with a single or multiple ACE history and not with our five MHP measures. Females, employed and respondents with physical illness participated significantly less often at all surveys than their counterparts (adjusted Odds Ratios (OR) are 0.79, 0.83 and 0.83 respectively). Married respondents and respondents of 35–49, 50–64 and 65 years old and older participated significantly more often than unmarried and 18–34 years old respondents respectively (adjusted ORs are 1.32, 1.95, 3.51, and 1.62 respectively). With respect to education, only those with an intermediate professional education level significantly more often participated at all surveys than respondents with a primary education or preparatory intermediate vocational education level (adjusted OR = 1.29).

Measures

Adverse childhood experiences

At T1, respondents were administered a modified version of the Negative Life Experiences and Trauma Questionnaire (NLETQ; Engelhard et al., 2003). It consisted of a list of 27 events (plus one category ‘no event’) and respondents were asked to rate whether they had experienced each event personally (0 = no, 1 = yes), and state the age(s) at which the event(s) took place (Hagenaars, 2016; Riem & Karreman, 2019). For this study, we first selected the following ACE-related items: (1) incest/sexual abuse as a child; (2) physical maltreatment as a child; and (3) psychological (emotional) maltreatment as a child. We then selected the items: (4) sexual violence (not rape) by a known person; (5) sexual violence (not rape) by an unknown person; (6) rape by a known person; (7) rape by an unknown person; (8) physical violence by a known person; (9) physical violence by an unknown person; (10) death of a sibling; and (11) death of a parent. Respondents who reported having experienced one or more abuse events during childhood (1–3) and/or other events (4–11) before the age of 18 were considered to have an ACE history.

Mental health problems (MHP)

Anxiety and depression symptoms were examined at T2 to T5 using the 5-item Mental Health Index or Inventory (Means-Christensen et al., 2005; Ware & Sherbourne, 1992). Respondents were asked to rate their mental health during the past month on a 6-point Likert scale (0 = never to 5 = continuously). After recoding the negative formulated items, the total scores were computed and multiplied by four (all Cronbach’s α > 0.85) according to the instructions of the MHI-5. Lower scores reflect a higher severity of MHP. Scores of 59 and lower (Driessen, 2011) were used to identify respondents with MHP.

Respondents were given a questionnaire listing 10 problems people may suffer from, varying from heart complaints to sleep problems. For this study, we focused on the items “Do you regularly suffer from fatigue?” and “Do you regularly suffer from sleep problems?” (0 = no, 1 = yes).

Mental health services use (MHS)

MHS use was assessed at T2 to T5 by means of two questions. The first question was “How often did you use the following health services over the past 12 months?”, with categories varying from family physician to alternative health practitioner. For the present study we focused on the use of a psychiatrist, psychologist or psychotherapist (0 = no, 1 = yes) in the past 12 months. In addition, respondents were asked a separate question: “Are you currently taking medicine at least once a week for anxiety or depression?” (0 = no, 1 = yes).

Physical illness

The presence of a medically diagnosed physical illness was assessed at T2–T5 by asking the question ‘Has a physician told you this last year that you suffer from one of the following (17) diseases/problems?’ (0 = no, 1 = yes). The conditions listed included “angina, pain in the chest; a heart attack including infarction or coronary thrombosis or another heart problem including heart failure’ and ‘cancer or malignant tumor, including leukemia or lymphoma’. Based on these items, we distinguished between respondents with and without one or more physical illness.

Demographics

Sex, age, being employed, education level and marital status were assessed at each survey.

Data analyses

To optimize the representativeness of our study sample, we first weighted the data using 32 exclusive demographic profiles among the total adult Dutch population in 2010 (N²⁰¹⁰ = 13,229,854 aged 18 and older), based on the register data of Statistics Netherlands. We focused on data for 2010 because the first health survey included questions that referred to the past year (2010–2011). The 32 profiles were constructed using the following demographic characteristics: sex (2 categories), age (4 categories), marital status (2 categories) and employment status (2 categories), totaling 2*4*2*2 = 32 exclusive profiles. All results are based on the weighted sample. Due to the weighting, the total number of subgroups may differ slightly between analyses. Chronic MHP was defined as having MHP at a minimum of three out of the four surveys taken (e.g. at T2, T3 and T5; at T2, T4 and T5; or at T2, T3, T4 and T5). For each MHP, the number of respondents with chronic MHP was calculated. Chronic or long-term contact with a psychiatrist, psychologist or psychotherapist and long-term use of medicines were calculated in a similar way as for MHP.

To answer the first and second research question, we used generalized estimating equations (GEE), a repeated measures logistic regression analyses, with MHP and MHS use at T2, T3, T4 and T5 as dependent variables. Time and group membership (1 = no-ACE, 2 = single ACE, 3 = multiple ACE) were entered as predictors. Since the study involved repeated measures over time, we used the auto-regressive correlation structure in GEE. In light of the possibility that the respondents’ employment status, level of education, marital status and physical illness might change over time, we used these variables as covariates at each corresponding survey (e.g. for the health variables assessed in November–December 2013, the demographic characteristics and illness assessed in November–December 2013 were used as covariates). We next examined differences in chronic MHP and MHS use, conducting multivariate logistic regression analyses with chronic MHP and MHS use as dependent variables, while controlling for sex, employment status, level of education, marital status, and physical illness at T2. Differences in the prevalence of different ACE between the two ACE subgroups were assessed using chi-square test.

To answer the third research question, four multivariate logistic regression analyses among respondents with chronic MHP were conducted with MHS as a dependent variable and group membership as a predictor. Respondents who used MHS and/or medicines for anxiety or depression symptoms at T2, T3, T4 and/or T5 were defined as MHS users. In the regression analyses, we controlled for demographics and physical illness at T2. Due to the cell counts, for all analyses we re-arranged education into three education levels (low = 1, 2; medium = 3; high = 4, 5). All analyses were conducted with IBM SPSS version 26.

Results

Respondent characteristics and Dutch population

The characteristics of the respondents without an ACE history (N = 2055, 84.7%), experiences with a single ACE (N = 280, 11.5%), and multiple ACE (N = 92, 3.8%) are presented in Table 1.

Table 1. Characteristics Dutch population and respondents.

	Dutch	T2 (2011)			T3 (2013)			T4 (2015)			T5 (2017)
	Popul. 2010	No- ACE	Single ACE	Multiple ACE	No- ACE	Single ACE	Multiple ACE	No- ACE	Single ACE	Multiple ACE	No- ACE	Single ACE	Multiple ACE
	%	%	%	%	%	%	%	%	%	%	%	%	%
Age
18–34 years	25.8	26.3	23.2	20.9	23.6	17.8	20.7	19.8	13.6	18.7	17.1	11.7	14.1
35–49 years	28.2	27.8	29.6	35.2	24.7	28.5	32.6	25.6	30.0	29.7	24.2	28.5	30.4
50–64 years	25.7	25.3	28.6	27.5	27.3	32.4	26.1	26.8	30.7	29.7	26.7	30.2	31.5
65 years or older	20.2	20.6	18.6	16.5	24.4	21.4	20.7	27.8	25.7	22.0	32.1	29.5	23.9
Sex
Man	49.1	50.5	41.8	40.2***	50.5	41.8	40.2***	50.5	41.8	40.2***	50.5	41.8	40.2***
Woman	50.9	49.5	58.2	59.8	49.5	58.2	59.8	49.5	58.2	59.8	49.5	58.2	59.8
Employed
No	45.8	45.7	43.9	55.4	47.7	44.1	56.5	46.6	50.0	66.3***	45.6	45.7	63.0***
Yes	54.2	54.3	56.1	44.6	52.3	55.9	43.5	53.4	50.0	33.7	54.4	54.3	37.0
Married
No	49.7	49.1	49.5	65.2**	48.9	47.0	62.0*	48.2	46.6	59.3	46.4	46.8	56.5
Yes	50.3	50.9	50.5	34.8	51.1	53.0	38.0	51.8	53.4	40.7	53.6	53.2	43.5
Education level
1 Prim. educ.	n.a.	26.8	27.8	19.6	25.8	27.1	19.6	25.1	26.8	17.6	24.7	27.0	17.6
2 Higher gen. sec.	n.a.	9.0	6.4	12.0	7.7	5.7	12.0	7.0	5.7	12.1	6.8	4.6	12.1
3 Inter. prof. educ.	n.a.	24.1	28.1	25.0	24.7	28.2	25.0	25.1	27.9	22.0	24.5	27.4	22.0
4 High. prof. educ.	n.a.	25.9	24.9	33.7	26.2	26.1	33.7	26.8	26.8	36.3	27.7	27.8	34.1
5 University	n.a.	14.2	12.8	9.8	15.6	12.9	9.8	16.1	12.9	12.1	16.4	13.2	14.3
Physical illness
No	n.a.	83.5	73.9	71.7***	80.7	72.2	69.6***	79.7	71.1	64.1***	78.0	71.4	60.4***
Yes	n.a.	16.5	26.1	28.3	19.3	27.8	30.4	20.3	28.9	35.9	22.0	28.6	39.6

ACE: confronted with one or more of the following adverse childhood experiences: sexual abuse; physical abuse; emotional abuse; sexual violence; rape; violence; and/or death of parent or sibling during childhood. Popul.: population. Education level: prim. educ: primary education, preparatory intermediate vocational education; higher gen. sec.: higher general secondary/pre-university educ; inter. prof. educ.: intermediate professional education; high. prof. educ: higher professional education. N.a.: not available for all Dutch residents. Results are based on weighted data and due to weighting numbers may slightly differ. n_{Single ACE} = 280. n_{Multiple ACE} = 92. n_no-ACE = 2055.

*p < 0.05. **p < 0.01. ***p < 0.001.

Prevalence ACE

Table 2 shows that the most common ACE in the single ACE group was the death of a parent/sibling, whereas the most common ACE in the multiple ACE group was emotional abuse.

Table 2. Adverse childhood experiences among single and multiple ACE group.

	Single ACE (n = 280)	Multiple ACE (n = 92)
	n (%)	n (%)
Incest/sexual abuse as a child	27 (9.6)	33 (35.9)
Physical maltreatment as a child	11 (3.9)	47 (51.1)
Psychological (emotional) maltreatment as a child	59 (21.0)	81 (88.0)
Sexual violence (not rape) by a known person^a	12 (4.3)	12 (13.2)
Sexual violence (not rape) by a unknown person^a	2 (0.7%)	6 (6.5)
Rape by a known person^a	3 (1.1)	4 (4.3)
Rape by a unknown person^a	2 (0.7)	5 (5.4)
Physical violence by a known person^a	13 (4.6)	16 (17.4)
Physical violence by a unknown person^a	17 (16.1)	4 (4.4)
Death of sibling/parent^a	134 (47.9)	14 (15.2)
	M (SD)	M (SD)
Number of ACE	1 (0)	2.42 (0.92)

Results are based on weighted data and due to weighting numbers may slightly differ.

^aBefore the age of 18.

MHP and MHS use

According to Table 3, a significantly higher proportion of adults with a single ACE and multiple ACE history suffered from assessed MHP compared to adults without an ACE history, and more often used MHS (contact with a psychiatrist, psychologist or psychotherapist) during the 6-year period. Respondents with multiple ACE compared to respondents with a single ACE significantly more often suffered from anxiety and depression symptoms (AOR = 1.90, 95% CI = 1.31–2.75, p = 0.001), fatigue (AOR = 1.79, 95% CI = 1.20–2.68, p = 0.002), sleep problems (AOR = 2.29, 95% CI = 1.54–3.43, p < 0.001) and more often used MHS (AOR = 1.80, 95% CI = 1.19–2.71, p < 0.001) during the 6-year period.

Table 3. Mental health problems and services use.

	Problems and services use
	2011 (T2)		2013 (T3)		2015 (T4)		2017 (T5)
	n	%	n	%	n	%	n	%	AOR^ace	(95% CI)	p-value
Anxiety and depression symptoms
No-ACE group (Ref.)	270	13.1	280	13.6	275	13.4	272	13.2
Single ACE group	61	21.8	65	23.2	64	22.9	48	17.1	1.62	(1.30- 2.01)	<0.001
Multiple ACE group	32	34.8	27	29.7	36	39.1	36	39.1	3.04	(2.22-4.24)	<0.001
Fatigue
No-ACE group (Ref.)	589	29.1	589	28.7	608	29.6	624	30.4
ACE group	98	35.0	111	39.6	119	42.5	113	40.4	1.41	(1.14-1.75)	0.002
Multiple ACE group	47	51.1	54	58.7	51	55.4	49	53.8	2.53	(1.77-3.63)	<0.001
Sleep problems
No-ACE group (Ref.)	367	18.0	384	18.7	402	19.6	414	20.2
Single ACE group	70	25.0	79	28.3	87	31.1	78	27.8	1.58	(1.26- 1.98)	<0.001
Multiple ACE group	43	46.7	49	53.3	45	48.9	46	50.0	3.61	(2.53-5.15)	<0.001
Contact with a psychiatrist, psychologist, or psychotherapist
No-ACE group (Ref.)	98	4.8	101	4.9	117	5.7	129	6.3
Single ACE group	24	8.6	36	13.0	24	8.6	23	8.2	1.72	(1.32- 2.25)	<0.001
Multiple ACE group	17	18.5	15	16.3	16	17.6	19	20.7	3.10	(2.17-4.41)	<0.001
Use medicine for anxiety or depression
No-ACE group (Ref.)	72	3.5	79	3.8	79	3.8	86	4.2
Single ACE group	12	4.3	11	3.93	14	5.0	14	5.0	1.13	(0.68- 1.87)	0.647
Multiple ACE group	7	7.8	7	7.6	10	10.9	9	9.9	2.20	(1.14-4.24)	0.019

ACE: confronted with one or more of the following ACEs: sexual abuse; physical abuse; emotional abuse; sexual violence; violence; rape; and/or death of parent or sibling during childhood. AOR^ACE: odds ratio (OR) of predictor ACE in repeated measures (GEE), adjusted for the covariates sex (2011), age, being employed, education level, marital status, and physical illness at each corresponding surveys. 95% CI: 95% confidence interval. Results are based on weighted data and due to weighting numbers may slightly differ. n^{Single ACE} = 280, n^{Multiple ACE} = 92, n^no-ACE = 2055.

Chronic MHP and MHS use

Table 4 demonstrates that a significantly higher proportion of adults with a single and multiple ACE suffered from assessed MHP compared to adults without an ACE history. A higher proportion of those with multiple ACE compared to those with an ACE history had ongoing contact with MHS. In addition, respondents with multiple ACE compared to single ACE more often suffered from chronic fatigue (AOR = 1.93, 95% CI = 1.17–3.19, p = 0.01) and sleep problems (AOR = 2.76, 95% CI = 1.63–4.67, p < 0.001), and more often used MHS (AOR = 4.66, 95% CI = 1.55–13.96, p = 0.005). The difference in anxiety and depression symptoms did not reach the p < 0.05 level (AOR = 1.82, 95% CI = 0.98–3.35, p = 0.056).

Table 4. Prevalence of chronic mental health problems and services use.

	Chronic problems and services use
	n^chronic MHP	%^chronic MHP	AOR^A^CE	(95% CI)	p-value
Chronic anxiety and depression symptoms
No-ACE group (Ref.)	154	7.5	1
Single ACE group	38	13.6	1.87	(1.26–2.78)	0.002
Multiple ACE group	24	26.1	3.40	(2.00–5.77)	<0.001
Chronic fatigue
No-ACE group (Ref.)	549	26.7	1
Single ACE group	105	37.5	1.46	(1.11–1.92)	0.007
Multiple ACE group	50	54.9	2.82	(1.80–4.40)	<0.001
Chronic sleep problems
No-ACE group (Ref.)	335	16.3	1
Single ACE group	71	25.4	1.50	(1.10–2.05)	0.010
Multiple ACE group	45	48.9	4.14	(2.62–6.55)	<0.001
Chronic contact with a psychiatrist, psychologist, or psychotherapist^a
No-ACE group (Ref.)	23	1.1	1
Single ACE group	6	2.1	1.68	(0.66–4.30)	0.278
Multiple ACE group	9	9.8	7.83	(3.45–17.80)	<0.001
Chronic use of medicine for anxiety or depression
No-ACE group (Ref.)	72	3.5	1
Single ACE group	9	3.2	0.77	(0.38–1.57)	0.472
Multiple ACE group	7	7.6	1.52	(0.66–3.51)	0.329

ACE: confronted with one or more of the following ACEs: sexual abuse; physical abuse; emotional abuse; sexual violence; rape; violence; and/or death of parent or sibling during childhood. Chronic = reported problem at 3 or 4 surveys. AOR^ACE: odds ratio (OR) of predictor ACE adjusted for the covariates sex, age, being employed, education level, marital status, and physical illness at T2. 95% CI: 95% confidence interval. Results are based on weighted data and due to weighting numbers may slightly differ. n^{Single ACE} = 280, n^{Multiple ACE} = 92, n^no-ACE = 2055.

^aDue to cell counts age was omitted from the list of predictors.

MHS use among those with chronic MHP

The prevalence of MHS use between T2 and T5 among respondents without an ACE history and chronic MHP did not differ significantly from respondents with a single ACE and a multiple ACE history who suffered from the same chronic MHP (MHS use^{chronic anxiety and depression} is 54.9, 55.3 and 50%, respectively; MHS use^{chronic fatigue} is 29.5, 36.8 and 44%, respectively, MHS use^{chronic sleep problems} is 31.3, 36.6, and 44.4%, respectively; full table is available upon request).

Test robustness of findings

About 60% of the respondents at T1 and T2 (2427/4126) participated in all five surveys. To assess the robustness of our findings, e.g. the extent to which results are biased because of the nonresponse, we used the following analytic strategy. First, using multiple imputation (MI, Rubin, 1987) we estimated missing data at T3, T4, and T5 using SPSS. We imputed total MHI-5 scores, and sex and age were computed using T2 data. With MI, we compiled three complete data sets (three samples with N = 4126) based on 20, 50, and 100 imputations because lower imputation numbers are discouraged (Graham et al., 2007). As in the analysis without imputations, we weighted the data. Next, we repeated all analyses for each data set separately. We finally compared the results based on the original study sample (N = 2427), with the results of the analyses of each of the three imputed data sets (based on pooled data). Findings showed similar patterns in results across the four samples: significant associations in one sample were significant in the other samples, and non-significant associations in one sample were non-significant in the other samples (the full tables are available upon request).

Discussion

To the best of our knowledge, this is the first multi-wave comparative population-based study on the effects of ACE on chronic problems and services use among those with chronic problems during 6-year period using a large probability sample. Results showed, in line with our expectations and previous studies (Brown et al., 1994; Collishaw et al., 2007; Hovens et al., 2012; McLaughlin et al., 2010; Suija et al., 2011), that adults with a single ACE and multiple ACE history significantly more often suffered from chronic MHP (chronic anxiety and depression symptoms, fatigue and sleep problems) during the 6-year study period than adults without an ACE history. Both ACE subgroups were also more often persistent users of MHSs. Previous research has shown that multiple ACE compared to single ACE is associated with a higher risk for MHP (Hughes et al., 2017; Selous et al., 2020) ; and findings showed that multiple ACE is also associated with a higher risk for chronic sleep problems and chronic fatigue.

We used a conservative criterion for chronic MHP. MHP had to be present at three or four of the four surveys (T2, T3, T4 and T5). We did not further examine chronic anxiety and depression symptoms levels among those with very low symptoms levels. Chronic fatigue and sleep problems were the most prevalent problems among both ACE subgroups, rather than the modest to severe anxiety and depression symptom levels that have received considerable attention in previous studies. These findings indicate that a focus on mental disorders such as major depression disorder underestimates the prevalence of MHP among adults with an ACE history.

The prevalence of chronic fatigue and sleep problems almost reached the point prevalence level of these problems at the four surveys among both the ACE and no-ACE subgroups, in contrast to anxiety and depression symptoms, and contact with mental health professionals. We are not aware of studies to compare these results with. Nevertheless, our results clearly suggest that when examining risk or protective factors for MHPs among adults with an ACE history, a distinction must be made between those with chronic problems and those with fluctuating problems.

As expected, and in line with the literature, adults of both ACE subgroups more often had contact with mental health professionals than adults without an ACE history (Bellis et al., 2017; Chartier et al., 2010; Larkin et al., 2018; Mills et al., 2012). A remarkable finding and in contrast to our expectations is that no differences were found between the prevalence of MHS use among adults with a multiple and single ACE history and chronic MHP, and no-ACE adults who also suffered from chronic MHP. Future research is warranted to gain insight in why an ACE history is not an extra barrier for MHS utilization among those with chronic MHPs. About half of the no-ACE and ACE subgroups with chronic anxiety and depression symptoms did not use MHS in the years before the health surveys indicating that attention and interventions from care providers and policymakers are needed to improve services use among these subgroups.

Limitations and strengths

An important issue in this field of research is that many studies, including our own, rely on recalled childhood experiences (Brewin & Andrews, 2017; Colman et al., 2016; Hardcastle et al., 2018). A recent meta-analysis comparing prospective and retrospective measures of childhood maltreatment within studies concluded that prospective and retrospective measures may identify different groups of individuals, given the low associations (about 50%) between prospective and retrospective measures (Baldwin et al., 2019). In addition, most studies on the effects of ACE on current mental health in adults assessed recalled ACEs and MHP concurrently. This may also have affected the associations reported, as current mental health status may bias recalled ACEs. However, in our study the first MHP assessments took place 6 months and the last MHP assessment took place more than 6 years after the ACE survey. In the longitudinal study by Colman et al. (2016), individuals who developed depression after approximately 11 years were more likely to report “new” ACE (or ACE they did not report previously). This pattern of reporting ACE was absent among those individuals with remitted depression or who had been consistently depressed. Another limitation of our study is that we did not conduct clinical interviews. Our study was conducted in the Netherlands: future studies should assess whether our findings are applicable to countries with other health care systems.

We have no information about other ACEs, such as food insecurity and homelessness (cf. Mersky et al., 2017). Our ACE measure partly differs from other ACE measures such as the Adverse Childhood Experiences-Questionnaire (Felitti et al., 1998) and Childhood Experience of Care and Abuse Interview (Bifulco et al., 1994), but include both threat and deprivation related items. Another aspect that remained unexplored was the developmental timing of ACE on the chronicity of MHP and MHS use (Riem et al., 2015). Our study was not aimed at the duration of ACE and mediators of the associations between ACE and mental health during adulthood (McLaughlin, 2016). It was outside the aim of the present study to examine differences between assessed types of ACE, such as between sexual and non-sexual related ACE.

The non-response was significantly associated with age, sex, employment, and marital status, but the final study sample was weighted for these demographic characteristics, and in all regression analyses we controlled for these variables. The results of the re-analyses after 20, 50 and 100 imputations of missing data showed identical patterns compared to the original study sample. We therefore assume that we minimized possible biased results due to the non-response to the extent possible.

Conclusion

Adults with a history of trauma-related ACE are at greater risk of suffering from chronic MHP. No evidence was found that adults with a single or multiple ACE history and chronic MHP found their way to MHS less often than adults without an ACE history with the same MHP. Nevertheless, improvement of services use is warranted because about 50% of those with chronic anxiety and depression symptoms did not use MHS (cf. Korotana et al., 2016).

Disclosure statement

No potential conflict of interest was reported by the author(s).