Abstract
Background:
Population-based studies have shown that less than one in two individuals reporting suicidal ideation also report past-year mental health service use. Only a few studies have looked at different types of providers consulted. There is a need to better understand the factors associated with different provider combinations of mental health service use in representative samples of individuals with suicidal ideation.
Aims:
The aim of the current study is to assess, using Andersen’s model of healthcare seeking behaviors, the predisposing, enabling and need factors associated with type of mental health service use in adults with past-year suicidal ideation.
Methods:
Data were drawn from the 2017 Health Barometer survey, a representative sample of the general population aged 18 to 75 years, among whom 1,128 respondents had reported suicidal ideation in the past year were analyzed. Past-year outpatient mental health service use (MHSU) was categorized into mutually exclusive groups as no use, general practitioner (GP) only; mental health professional (MHP) only; and both GP and MHP. Multinomial regression analyses were used to model mental health service use as a function of predisposing, enabling and need factors.
Results:
Overall, 44.3% reported past-year MHSU and this was higher in females than males (49.0% vs. 37.6%). Prevalence of GP only use in the overall sample was 8.7%, consulting with GP and MHP was 21.3%, consulting with MHP only was 14.3%. Higher education was associated with increased MHP use. Residing in a rural area was associated with increased GP only use. Presence of a suicide attempt within the year, a major depressive episode and role impairment were associated with consulting a GP and MHP, and MHP only, but not GP only.
Conclusions:
When controlling for need and predisposing factors, socio-economic factors related to employment and income were associated with higher levels of consulting with mental health professionals.
Introduction
Suicide is an important worldwide public health issue (World Health Organization, 2017) as it is also one of the most preventable causes of premature death. Large population studies have shown that 60% of individuals reporting suicidal ideation also report a suicide plan and attempt within the first year (Harmer et al., 2020; Nock et al., 2008), which calls for increased attention on suicidal thoughts and behaviors (Obegi, 2019). Suicidal ideation as a symptom of other primary psychiatric diagnoses, mainly mood disorders highlight the importance of early and accessible suicide prevention interventions in primary care as the majority of suicide decedents consult in primary care (Stene-Larsen & Reneflot, 2019). Evidence-based suicide prevention strategies (Mann et al., 2021) include among others, educating general practitioners on screening and effective treatment of depression and known effective pharmacotherapy and psychotherapy (American psychiatric Association, 2021; Kennedy et al., 2016; MacQueen et al., 2016; National Institute for Health and Care Excellence, 2011; Parikh et al., 2016) such as cognitive behavioral therapy.
Population-based studies have been consistent in showing that up to one in two individuals with suicidal ideation will report past-year mental health service use (Bruffaerts et al., 2011, 2019; Cheung et al., 2009; Stanley et al., 2015; Vasiliadis et al., 2022). When looking at type of health care providers consulted, only a handful of population-based studies have addressed this issue. In individuals aged 15 years and over reporting suicidality (e.g. ideation and attempts) in the 2002 Canadian Community Health Survey (CCHS 1.2), past-year prevalence of outpatient consultations with psychiatrists ranged between 6.6% and 20.0%, consultations in primary medical care ranged between 11.7% and 24.5%, and consultations with other mental health providers ranged between 5.8% and 7.9%, (Rhodes et al., 2006). Others focusing on adolescent (15–18 years) and young adult (19–24 years) males and females reporting suicidality (e.g. ideation and attempts), past-year mental health service use prevalence estimates ranged between 17.5% and 23.3% for consulting with psychiatrists, 14.7% and 41.3% for consulting with general practitioners, 21.0% and 24.6% for consulting with psychologists, 12.8% and 31.9% for consulting with social workers/counselors, and 9.3% and 16.2% for consulting with other providers (Cheung & Dewa, 2007).
Studies on the factors associated with different provider types have not been consistent with respect to age and gender (Batterham et al., 2022; Cheung & Dewa, 2007; Mok et al., 2021). Rare are the studies reporting on socio-economic factors. Education was associated with psychiatrist or psychologist service use but not with inpatient or outpatient professional service use in general, while employment status and region of residence were not (Batterham et al., 2022).
More recent studies with representative samples of individuals with suicidal ideation are needed to better understand the factors associated with different provider combinations of mental health service use including general practitioners, family physicians, and mental health specialists which may be associated with increased efficiency of the health system. A Canadian study showed increased patient and physician satisfaction when physicians were able to refer patients to psychologists as well as improved access to mental health treatment and optimal use of physician time (Chomienne et al., 2011). Socio-economic inequities in accessing mental health professionals, but not general practitioners, in primary care have been reported (Vasiliadis et al., 2009).
In a French representative general population sample of adults reporting suicidal ideation, the aim of the current study is to examine, according to Andersen’s model (Andersen, 1995) of healthcare seeking behavior, the predisposing (e.g. sociodemographic), enabling (e.g. economic, contextual), and need (e.g. health status) factors associated with consulting in the past-year a general practitioner only, mental health professional only, and both general practitioner and mental health professional. This study will inform on the presence of inequities regarding mental health service use in individuals with suicide ideation, which may contribute to health policies that respond to the mental health needs of individuals with suicidal ideation and this irrespective of socio-economic status.
Methods
Survey sample
This study relied on data drawn from a representative general population sample of adults aged between 18 and 75 years responding to the 2017 Santé Publique France Health Barometer survey on health behaviors, mental health and substance use, and health care services. Additional information regarding sampling strategy and survey methodology has been previously published (Richard et al., 2018). Study participants were recruited following a two-stage random sample of residents living in metropolitan France. A random sample of phone numbers was generated for each household, where subsequently one resident was randomly selected. In addition, mobile phone numbers were randomly generated to reach individuals who may or may not have a landline. If a household or respondent refused or could not be contacted, there was no replacement. The survey was conducted in French only, speaking French was therefore a requirement for participation. The health survey was carried out between January and July 2017 by trained professional interviewers using a computer-assisted telephone interview lasting on average close to 30 min. Study approval was obtained by the French commission on data privacy and public liberties (CNIL). Informed verbal consent was obtained from each study participant. The overall survey response rate was 48.5%, and yielded a total of 25,319 respondents. The presence of past-year suicidal ideation ‘Over the past 12 months, have you thought about suicide?’ was reported by 4.7% (n = 1,148) of respondents. The current study sample includes individuals with complete data (n = 1,128).
Measures
Dependent variable of interest
Past year health service use for mental health reasons
The dependent variable of interest, general and specialist mental health service use for mental health reasons, was ascertained with the following question: « In the past 12 months, have you used the services of an organization such as a hospital, clinic, center, hotline, or consulted a professional such as a doctor, psychiatrist, therapist, for emotional, nervous, psychological, or behavioral problems? ». Respondents answering yes to this question, were then asked which type of services they had used and the following choices of answers were considered for analyses: general practitioner or a family doctor (GP); psychiatrist; psychologist; don’t know whether psychiatrist or psychologist; other professional (open ended question, which included responses such as psychotherapist, psychoanalyst, hypnotist, acupuncturist, sophrologist). Health service use for mental health reasons was defined according to combinations of professionals consulted: no health service use, general practitioner (GP) only, psychiatrist or psychologist or other professional (MHP) only, GP and MHP.
Independent variable of interest
Predisposing factors
The predisposing study factors included the following: sex (female, male), age (categorized in four groups: 18–24, 25–34, 35–64, 65–75 years), sexual orientation (heterosexual, non-heterosexual/undisclosed), marital status (categorized according to two groups: single/separated/divorced/widowed/other/undisclosed vs. married/living with partner), highest degree attained in education (less than high school, high school or vocational degree and at least some university) and born outside of metropolitan France (yes, no).
Enabling factors
The enabling study factors included the following: urbanicity of region of residence (rural vs. urban community), employment status (in employment, apprentice/paid internship or student/in training, unemployed, retired or in pre-retirement, stays at home, other (long term leave, person with disability), additional complementary medical insurance (yes, no). Income was based on the adjusted household disposable income and assessed as the net monthly total income and number of individuals living in household (https://data.oecd.org/hha/household-disposable-income.htm), which was categorized in terciles. Missing data on income was studied as a separate category.
Need factors
The need study factors included a number of clinical factors, role impairment and substance use. The presence of a suicide attempt, assessed with the following « ‘During your life, have you attempted suicide’ and ‘Did this attempt occur over the past 12 months?’ », was categorized as no lifetime attempt, past-year attempt and attempt over a year ago. The presence of a major depressive episode in the past year was assessed with World Health Organization’s Composite International Diagnostic Interview – Short Form (CIDI-SF) (Kessler et al., 1998). Self-rated general health was categorized as very good/good versus fairly good/poor/very bad. Role impairment due to a health problem for at least 6 months was assessed according to three options: yes, severely; yes, moderately; none, no impairment at all. Self-reported number of times drunk was categorized as ⩾10 times in the past year, yes/no. Self-reported past-year illicit drug use was also assessed (e.g. hallucinogenic mushrooms, poppers, other inhalants such as glues or solvents, ecstasy or MDMA, amphetamines, LSD, cocaine, crack or freebase, heroin, Subutex, methadone, GHB or GBL, Purple Drank or DXM or other). Cannabis use disorder risk was assessed with the 6-item cannabis abuse screening with scores ranging from 0 to 24 (Legleye et al., 2007) and categorized as no or low risk (scores 0–2), moderate (scores 3–6) and severe risk (scores ⩾7) (Legleye et al., 2013, 2015). Cannabis and illicit drug use were not assessed in older adults aged ⩾65 years. Studies have shown the rarity of illicit drug use and abuse and dependence in older adults (Blazer & Wu, 2009). Missing information on illicit drug and cannabis use disorder were therefore considered as absent in the present analyses.
Data analysis
Descriptive statistics presented weighted estimates (proportions and standard errors). Multinomial regression analyses were carried out to study type of health service use for mental health reasons as a function of predisposing, enabling and need factors. Odds ratios (OR) and 95% confidence intervals (CI) presented were weighted. Analyses were performed using SAS version 9.4.
Results
In the current sample, 44.3% of respondents with suicidal ideation reported consulting with either a GP or MHP for mental health reasons in the past year (Table 1) and this was higher in females than males (49.0% vs. 37.6%). GP were the most widely consulted (30.0%) followed by psychiatrists (22.4%), psychologists (20.0%) and other health professionals (8.7%). Females were almost twice more likely than males to consult a GP (OR 1.96; 95% CI [1.50, 2.56]) and more likely to consult a psychiatrist (OR 1.55; 95% CI [1.16, 2.07]) but not a psychologist (OR 1.30; 95% CI [0.97, 1.75]). When looking at the combinations of type of professionals consulted, GP only use in the overall sample was 8.7%, and higher in females (10.1%) than males (6.7%) (OR 1.55; 95% CI [1.00, 2.40]); GP and MHP in the overall sample was 21.3% and was higher in females (25.6%) than males (15.3%) (OR 1.91; 95% CI [1.41, 2.58]); and MHP only in the overall sample was 14.3% with no significant difference between females (13.4%) and males (15.5%) (OR 0.84; 95% CI [0.60, 1.17]).
Type of health service use for mental health reasons among adults with suicidal ideation in the past year, overall and by sex.
Note. Estimates presented are weighted, (Overall: N = 1,168, weighted; Female: N = 687; Male: N = 481).
Mental health professional only includes consultation with psychiatrist, psychologist or other health professional for mental health reasons.
The multinomial analyses are presented in Table 2. Among the predisposing factors studied, females were more likely than males to consult a GP only and a GP and MHP. Respondents reporting the use of GP and MHP services, and MHP services only, were more likely to report attaining high school or vocational degree (aOR 1.85, 95% CI [1.15, 2.99]; aOR 1.89, 95% CI [1.14, 3.12]) and at least some university (aOR 2.92, 95% CI [1.95, 4.35]; aOR 1.85, 95% CI [1.16, 2.97]).
Multinomial analyses of type of mental health service use in individuals with past-year suicidal ideation.
Note. Estimates presented are weighted. ‘-’ Not reported, confidence interval too wide.
Bold signifies p < .05.
Mental health professional only includes consultation with psychiatrist, psychologist or other health professional for mental health reasons.
Among the enabling factors, residing in a rural as compared to an urban area was associated with increased GP only use (aOR 1.85; 95% CI [1.14, 2.99]). In regard to employment, individuals reporting GP only were less likely to be on long-term disability than those employed (aOR 0.34; 95% CI [0.13, 0.88]), reporting GP and MHP were less likely to report staying at-home (aOR 0.30; 95% CI [0.10, 0.91]), while those reporting MHP only were more likely to be unemployed (aOR 2.05; 95% CI [1.21, 4.22]) and report staying at-home (aOR 2.59; 95% CI [1.11, 6.03]). Reporting income in the second (aOR 1.81; 95% CI [1.19, 2.76]) and third (aOR 2.43; 95% CI [1.39, 4.22]) income terciles were associated with consulting a GP and MHP, and MHP only.
Among the need factors studied, reporting a suicide attempt within the past year was associated with consulting a GP and MHP (aOR 3.33; 95% CI [1.73, 6.40]) and MHP only (aOR 3.28; 95% CI [1.58, 6.79]). Reporting a past-year major depressive episode was also associated with consulting a GP and MHP (aOR 3.09; 95% CI [2.19, 4.34]) and MHP only (aOR 2.16; 95% CI [1.48, 3.16]). The presence of role impairment was also associated with GP and MHP, and MHP only. Severe risk of cannabis use disorder was associated with increased use of GP only (aOR 2.61; 95% CI [1.07, 6.36]).
Discussion
The current study contributes to the literature by presenting in a French population of adults and older adults reporting suicidal ideation the predisposing, enabling and need factors associated with mental health service use of different type of health professional combinations. The main findings showed that among those reporting mental health service use, a higher proportion reported consulting a GP and MHP, followed by MHP only and GP only. Differences in predisposing, enabling and need factors were observed in the type of health providers consulted.
France has a universal health protection law with statutory and voluntary health insurance plans (Durand-Zaleski, 2020) where, although there is a voluntary gatekeeper system for those aged 16 years and over, 90.9% have a chosen gatekeeper, among which in 95% of cases is a GP. Self-referrals to psychiatrists are allowed. The majority of mental health disorders are treated by GPs, psychiatrists and psychologists in the private sector on an outpatient basis. Payments for consultation and health services are made directly to health providers. Patients then file insurance claims and are reimbursed depending on their coverage and copayments.
The study findings showed differences in the predisposing, enabling and need factors associated with the different combination types of health providers consulted in the past-year. Females were more likely than males to have consulted a GP only and GP and MHP as compared to no mental health service use. Others researchers have similarly reported a higher likelihood of females with suicidality consulting a GP as compared to no health service use (Mok et al., 2021), and no difference between males and females with suicidality consulting mental health professionals as compared no use (Batterham et al., 2022; Mok et al., 2021). In one of the scarce studies on the factors associated with consulting different providers in individuals with major depression using health services, the presence of suicidal ideation was not associated with consulting a GP/MHP as compared to GP only, nor MHP only as compared to GP only (Gagné et al., 2014). Of interest, after controlling for need and enabling factors, country of birth, age, and sexual orientation were not associated with any type of service use in this French population sample of individuals with suicidal ideation. This is of great importance to underline, which suggests the lack of socioeconomic disparities in regard to these factors (Bourgois et al., 2017).
As compared to respondents reporting less than a high school education, those with a higher level of education were more likely to consult a GP and MH, and MHP only, but not GP only. Previous research on the large Dutch NEMESIS similarly showed that higher education was associated with less primary care only use for mental health reasons with higher likelihood of mental health care in adults (ten Have, Have et al., 2003). Higher education has been shown to be associated with better attitudes toward mental health treatment and associated with use of mental health professionals (Gonzalez et al., 2011; Leaf et al., 1987; ten Have et al., 2003).
The current findings also showed that among the enabling factors, region of residence, employment and income were associated with different types of health service use. The study showed that as compared to individuals with suicidal ideation not seeking services, individuals living in rural areas were more likely to consult a GP only. No association was observed between region of residence and GP/MHP and MHP, and MHP only as compared to no use. Others have similarly shown no significant relationship between any health service use and mental health professional use and rural, regional or metropolitan location of residence (Batterham et al., 2020, 2022). As compared to individuals who were employed, those who were unemployed and stayed at home were more likely to report the use of a mental health professionals only. These findings may reflect the association between psychiatric clinical severity and not being employed (Bourgois et al., 2017; Defebvre & Barnay, 2016). Higher income was also associated with consulting MHP overall, which has been similarly reported in other public health systems (Vasiliadis et al., 2009). As individuals have to pay out of pocket at the point of medical contact in France, income may play a role as an affordability barrier in seeking specialized mental health care.
When controlling for predisposing and enabling factors, the presence of major depression, a suicide attempt within the past year, and role impairment were each associated with specialized mental health services with and without consulting a GP, which has been reported elsewhere (Batterham et al., 2022). Individuals with suicidal ideation reporting severe risk for cannabis use disorder were more likely to consult with a GP only, while no difference was observed for consulting with MHP. Although the increased health service use in general medical services is a positive outcome, the risk of severe cannabis use disorder in primary care populations with suicidal ideation needs further study as these individuals with suicidal ideation may be using cannabis as a coping mechanism for underlying psychiatric symptoms needing specialized mental health and addiction services.
The study findings should be considered with the following limitations. First, the analyses were based on data collected from the 2017 France Health Barometer survey and findings can be generalizable to adults and older adults with suicidal ideation living in the community and covered under national public healthcare plans. Information on the presence of suicidal ideation and health service use for mental health reasons are based on self-reports and may be subject to social desirability and recall bias. The use of health services was for mental health reasons and not specifically for suicidal ideation. The findings of the study cannot conclude on the quality of mental health treatment received nor as to referral patterns between GP and mental health professionals or collaborative care, nor between inpatient and ambulatory visits.
In conclusion, more than one in two individuals reporting suicidal ideation did not consult mental health services in the past-year suggesting potential unmet mental health service needs. Given France’s public health system, it was important to note that when controlling for need and enabling factors, socio-demographic factors such as age, marital status, being born outside of France and sexual orientation were not associated with mental health service use. Given the pandemic context and stressors related to confinement, future research could also focus on whether individuals lived alone and family structure (e.g. single-parent families) are related to mental health service use. Socio-economic factors including education, employment and income were however associated with MHP use after controlling for need and predisposing factors. Income-related inequities may be present in accessing mental health specialists. Future research should focus on detailing the type of mental health treatment received and adequacy of mental health care in individuals reporting suicidal ideation.
Finally, the findings help inform on the importance of public health policies aimed at decreasing regional, educational and financial related barriers in accessing specialist mental health service providers. Policies where individuals do not need to cover mental health consultation fees up front may help in increasing access to mental health specialists. Improved integrated care between general practitioners and mental health specialists, including follow-up with referrals, may also help in removing silos and increasing continuity and accessibility to mental health specialists among individuals with less education and living in rural areas. In line with these principles, a recent policy has been put in place in France whereby up to eight visits per year with a psychologist can now be reimbursed when prescribed by the general practitioner.
Footnotes
Declaration of Conflict of interest
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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Helen-Maria Vasiliadis was a visiting scholar, funded by the Initiative d’Excellence (IdEx) of Université de Bordeaux, at the time of the study. The funder had no role in collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.
Data availability
The authors are not legally authorized to share or publicly publish the dataset. Requests for access to the data should be addressed to Santé publique France.