Problematic substance use: an assessment of workplace implications in midwifery

Abstract

Background

Problematic substance use (PSU) poses occupational, personal and professional risks. As an occupational group, midwives have been under-represented in research on PSU.

Aims

The aim of this study was to assess self-reported occurrences of PSU, help-seeking behaviours and barriers, and perceptions of impairment in UK-based midwives.

Methods

Self-selecting registered midwives were anonymously surveyed using the Tobacco, Alcohol, Prescription Medications, and Substance Use/Misuse (TAPS) tool, the Perceptions of Nursing Impairment Inventory (PNII) and open-ended/closed questions. Quantitative data were used to explore PSU, help-seeking and attitudes to impairment. Qualitative responses were used to provide richer understandings.

Results

From 623 completed surveys, 28% (n = 176) self-reported PSU in response to work-related stress and anxiety, bullying, traumatic clinical incidents and maintenance of overall functioning. PSU was related to alcohol and a range of restricted drugs. While 11% of those affected indicated they had sought help, 27% felt they should seek help but did not. Reported barriers to help-seeking included fear of repercussions, shame, stigma, practicalities and a perceived lack of support either available or required. Perceptions of impairment were predominantly compassionate with a minority of stigmatizing attitudes displayed.

Conclusions

Overall, 10% of the sample reported they had attended work under the influence of alcohol, and 6% under the influence of drugs other than tobacco or those as prescribed to them. Furthermore, 37% indicated concern about a colleague’s substance use. As stigmatizing attitudes and punitive actions can dissuade help-seeking, changed perceptions and policies which favour alternatives to discipline are suggested to reduce the risk overall.

Fitness to practise, midwifery, occupational health, substance use

Key learning points

What is already known about this subject:

• Between 2014 and 2016, the Nursing and Midwifery Council saw fitness to practise cases related to alcohol (16%) or drug consumption (10%).

• Barriers to help-seeking for those with problematic substance use include stigma.

• Problematic substance use has been studied in nurses and physicians, but a lack of evidence exists for midwifery.

What this study adds:

• Self-reported problematic substance use was 28% (n = 176), of whom 11% (n = 20) reported seeking help and 27% (n = 47) did not seek help when needed.

• Work-based issues influenced problematic substance use with profession-related issues indicated as barriers to help-seeking, alongside stigma.

• Perceptions of impaired midwives were broadly compassionate. However, stigmatizing views can compound barriers to help-seeking.

What impact this may have on practice or policy:

• A potentially enhanced risk of problematic substance use in midwives compared to allied professions requires a greater understanding of the underlying occupational issues.

• Compassionate, de-stigmatizing interventions supporting midwives with problematic substance use may overcome barriers to help-seeking.

• Policies favouring a non-punitive approach may be advantageous for promoting a sustainable, safer and healthier midwifery workforce.

Introduction

In a recent survey of midwives based in the UK [1], 1464 of 1997 respondents (83%) reported episodes of work-related stress and burnout. Evidence suggests that substance use may occur in healthcare professionals due to such episodes [2], leaving them depleted [3]. Problematic substance use (PSU) in such populations may impact upon performance, relationships, attendance, reliability and the quality and safety of care given [3,4]. From a total of 1298 fitness to practise cases put before the Nursing and Midwifery Council (NMC) in the UK between 2014 and 2016, 16% (n = 208) related to impairment due to alcohol and 10% (n = 131) related to impairment through drug use [3]. While there is no definition of ‘impairment’ provided by the NMC’s legislative framework, and there are some subtle and complex philosophical differences in how it is conceptualized, in this professional context, impairment may be generally referred to as the inability to practise with adequate professionalism, skill and/or safety due to the use of alcohol and/or drugs.

While recent studies have focussed upon PSU in nursing populations [5], an integrative systematic review of the literature has revealed a paucity of evidence on PSU in midwifery populations [6]. The increased risk to health and impaired functioning in healthcare workers with PSU [7] suggests that if PSU is a significant issue within midwifery, it could be an additional factor to consider in addressing current midwifery workforce challenges [8]. Midwives and nurses who become impaired through PSU may have conditions placed upon their practice, be suspended or removed from the NMC register [9]. This, along with the stigma associated with a perceived ‘failure to cope’ also leaves many reluctant to seek help [4,10]. This is concerning, because as is the case in other professional groups [4], delays in receiving treatment may prolong impairment, and thus the associated risks to both professionals and the public [11]. With little evidence of the scale of self-reported occurrences of PSU in midwives based within the UK, other than from those referred to the NMC [3], it is not yet possible to assess its potential impact on both the profession, individual midwives and the public. Therefore, we distributed an anonymous self-selecting survey to UK midwifery networks to identify self-reported incidents of substance use in UK midwives, understand help-seeking behaviours and barriers to help-seeking, and explore midwives’ perceptions towards colleagues impaired by PSU.

Methods

We employed a confidential and anonymous self-administered mixed-method online survey, as these are evidenced to encourage participants to disclose potentially socially undesirable behaviours in relation to PSU [12]. Recruitment began after ethical approval was granted from the appropriate ethical review committee. To be eligible, participants were required to be over the age of 18 years and registered as a midwife with the NMC. They self-selected to contribute by responding to an advert placed in an editorial of the British Journal of Midwifery and shared by the Royal College of Midwives, the research teams’ professional networks and the ‘Make Birth Better’ network online. Based on the reported prevalence of PSU in a nursing population [4], we used a proportion estimate of 0.2 with 2% margin of error. With a population of 45 060 NMC registered midwives and dual nurse/midwives, we calculated a sample size of 457 appropriate to estimate a simple proportion of PSU within UK midwives. However, due to the nature of the online survey, an exact response rate is not possible to determine [13]. Data collection began in January 2020 and ended on 20 March 2020. Therefore, data were collected prior to the COVID-19 lockdown in the UK.

Demographic questions were related to age, gender, education and employment. Other variables included measures of substance use and perceptions of impairment. Two standardized instruments with established validity and reliability provided the foundation of the survey: Tobacco, Alcohol, Prescription Medications, and Substance Use/Misuse (TAPS) tool [14] and the Perceptions of Nursing Impairment Inventory (PNII) [15]. PNII statements were adapted for a midwifery cohort based in the UK. A subset of questions were used to capture help-seeking behaviours, attending work under the influence of drugs and/or alcohol and concern for colleagues’ PSU. Open-ended qualitative responses were invited to provide annotation to the answers given, including reasons for engagement in PSU. No timeframe restrictions were placed upon responses to open-ended questions; therefore, participants were able to share and recall incidents spanning their whole careers.

Responses to the two-step TAPS tool were analysed in line with the benchmark for diagnosing PSU [16]. Questions within TAPS-1 related to frequency of tobacco, alcohol above recommended daily limits (more than five drinks per day for men and more than four drinks per day for women), illicit drugs and non-medical use of prescription medications (sedatives, opioids, and stimulants) use in the past 12 months. Any participant who indicated anything other than a negative response on the TAPS-1 screening tool was then assessed for problematic use via the TAPS-2 screening tool. Questions within TAPS-2 assessed use of tobacco, alcohol, six different classes of illicit drugs and other drugs during the past 3 months. For tobacco and other regulated drugs, a score of 1+ was set as the cut off. For alcohol, we equated a score of 1 to ‘use’, but not ‘problematic use’, thus a cut off of 2+ was used for diagnosis of PSU in line with current recommendations [14].

Responses to the PNII were analysed by the proportion of positive responses per item (agree or strongly agree), broadly in line with previous uses of the scale in a nursing population [17]. We analysed the qualitative responses for each open-ended question using qualitative content analysis [18]. Statements were classified into a number of categories which represented a similar sentiment. These categories were then assimilated into themes broadly representing an over-arching meaning. The number of statements related to each theme have been counted to provide an illustration of the salience of each theme within the sample (see Table S1, available as Supplementary data at Occupational Medicine Online). We chose illustrative quotes which best represented the overall sentiment of the data within each theme. We were unable to present some of the qualitative data as quotes as they included descriptive accounts of particularly unique events and thus posed a threat to anonymity.

Results

All 623 completed returns were included in the analyses. To assess broad representativeness, the study sample demographics have been compared to the study population (Table 1). The number of male respondents in the sample is broadly representative of the study population but the low number (n = 3) coupled with low numbers of male midwives in practice creates a risk to anonymity. Findings are therefore presented for the combined sample and all analyses conducted with the sample as a whole. The spread of age for the sample is younger than that in the overall population with a greater proportion of younger midwives responding to the survey which has the potential to introduce bias in the results. Overall, 64% (n = 397) participants were in full-time employment, with 32% (n = 200) in part-time employment. Only 2% (n = 13) of participants indicated that they were either agency or bank staff and 2% (n = 13) indicated that they were not currently employed. No comparable data are freely available on employment status of the population as the NMC aggregates these data with nurses.

Table 1.

Comparison of study population (NMC, 2018–19) with study sample

	Population (%)	Sample (%)
Gender
Female	>99	>99
Male	<1	<1
Age
<30	19	25
31–40	27	32
41–50	23	24
>51	31	19

Self-reported PSU within the sample across all substances was 28% (n = 176) with alcohol use disorders most common at 16% (n = 101), and 6% (n = 37) of respondents having a positive screen for multiple substances. Within this analysis, tobacco is analysed as a freely available substance distinct from regulated drugs, with 8% (n = 50) reporting a dependence on smoking tobacco. Self-identified PSU by demographic variable is displayed in Table 2.

Table 2.

Self-identified PSU by age and employment status (out of 623 responses)

	PSU	Problematic alcohol use	Problematic tobacco use	Problematic druga use
	n (%)	n (%)	n (%)	n (%)
Total	176 (28)	101 (16)	50 (8)	67 (11)
Age
<30	44 (28)	19 (12)	13 (8)	20 (13)
31–40	73 (36)	48 (24)	19 (10)	31 (15)
41–50	39 (27)	25 (17)	14 (10)	9 (6)
>51	20 (17)	9 (8)	4 (3)	7 (6)
Employment
Full time	119 (30)	68 (17)	33 (8])	46 (12)
Part time	51 (26)	28 (14)	17 (14)	20 (10)
Agency/bank	3 (23)	3 (23)	0 (0)	0 (0)
Not employed	3 (23)	2 (15)	0 (0)	1 (8)

^aIncludes both prescribed and illicit drugs.

Self-reported problematic drug use within the sample was 11% (n = 67) with sedatives the most common drug used by 6% (n = 68) of the total sample. A breakdown of the self-reported use of each drug type can be found in Table 3. Within this sample, 72% (n = 447) of respondents had a negative screen for PSU.

Table 3.

Self-reported problematic usage by substance (out of 623 responses)

Substance	n (%)
Alcohol	101 (16)
Smoking	50 (8)
Cannabis	30 (5)
Stimulants	12 (2)
Heroin	3 (<1)
Opioids	14 (2)
Sedatives	68 (6)
Medication for the management of ADHD (for example, Adderall or Ritalin)	2 (<1)

Thirty-six respondents reported problematic use across multiple substances. (PSU was defined by a score of 1+ for drugs and tobacco and 2+ for alcohol in responses to the TAPS-2 screening tool.)

When asked about attending work under the influence, 10% (n = 62) of respondents indicated they had ‘attended work under the influence of alcohol’ and 6% (n = 36) indicated they had ‘attended work under the influence of drugs other than tobacco or those as prescribed to them’. While there remains a lack of agreement on the operational definition of impairment in this context, PNII statements most widely agreed upon were that the regulator should provide midwives suspected of impairment information regarding their rights in any disciplinary process, and that employee assistance programmes should be a requirement for support (see Table 4). Statements least agreed upon were the suggestion that impairment in relation to alcohol and/or drugs is due to a personality weakness, and that impaired midwives could not be productive or trustworthy after treatment.

Table 4.

Perceptions of impairment ranked by percentage of positive responses (agree or strongly agree)

	Agree
	n (%)
The Regulator’s responsibility should include offering the impaired midwife referral to sources of assistance	596 (96)
Major healthcare agencies should be required to provide employee assistance programs which could serve the impaired midwife.	589 (95)
Midwives have an obligation to notify their manager when they suspect impairment in a co-worker	550 (88)
Public safety can be assured by placing a caution period on the registration of the impaired midwife	510 (82)
If an impaired midwife is receiving treatment, it is important for his/her manager and co-worker to be aware of the fact as they are usually able to offer assistance and /or help them to receive assistance	476 (76)
When a manager has concrete evidence that a midwife is impaired, the manager has a responsibility to suspend that individual pending investigation of the charges	433 (70)
Impaired midwives can best be understood as people who suffer from an illness.	365 (59)
When suspecting impairment in a co-worker, the midwife’s first response should be to confront the individual	300 (48)
When a manager has concrete evidence that a midwife is impaired, the manager has a responsibility to dismiss that individual immediately and report the case to my regulatory body	151 (24)
In most cases, public safety should require that the impaired midwives’ registration be revoked	100 (16)
For purposes of public protection, the regulator should publish the names of all midwives found to be impaired	46 (7)
Even after treatment it is unusual for an impaired midwife to be productive, trustworthy, and capable of working as a registered midwife.	42 (7)
Impairment is generally the result of a weakness in the midwife’s personality.	23 (4)

When participants were asked why they ‘typically use substances not as prescribed, illegally or that were not prescribed’ for them, the statements offered (n = 33) were categorized into themes of work-related stress and anxiety, bullying, traumatic clinical incidents and the maintenance of overall functioning. Here, there were a particular spectrum of statements categorized into themes of work-related stress and anxiety, where some participants broadly described letting their hair down after a challenging shift, where others broadly described grappling with burnout. When asked whether they were ‘concerned about a colleague in relation to their use of substances’, 37% (n = 229) of respondents indicated positively. Statements offered in relation to the nature of their concern and the substance, circumstances and outcomes associated with it (n = 200) were categorized into themes of impairment through problematic alcohol use and addiction, problematic opioid use, lack of compassion toward problematic nitrous oxide (Entonox) use, functioning with cocaine, problematic pain medication use, problematic use of sleeping aids and problematic cannabis use.

Of those midwives who met the criteria for PSU, 11% (n = 20) indicated they had sought help, and 27% (n = 47) indicated they felt they should seek support but did not. When participants were asked why they had not sought help, statements offered in relation to this question (n = 108) were categorized into themes relating to fear of repercussions, shame and stigma, support perceived to be unavailable, help not wanted or perceived as not required and perceived impracticalities. When participants were invited to offer any further information, some statements were categorized as stigmatizing perceptions (n = 7), yet the majority were categorized as compassionate (n = 26). In respect of all open-ended questions, illustrative quotes are presented for each theme alongside the number of statements offered for each in Table S1 (available as Supplementary data at Occupational Medicine Online).

Discussion

Self-reported PSU across all substances was identified in just under a third of the midwives participating in this sample (28%); higher than that reported in nursing (8–20%) [5] and physician populations (8–15%) [19]. PSU within the present sample was reported in relation to alcohol, cannabis, nitrous oxide, cocaine, cannabis, amphetamines, pain medication, heroin, sedatives and/or medication for the management of Attention deficit hyperactivity disorder (ADHD). The percentage of midwives who met the criteria here for alcohol use disorder (16%) is higher than the percentages previously found for alcohol use disorders in nursing populations (6–10%) [20]. The percentage of midwives within the present sample reporting cannabis use (5%) lies marginally higher than the global usage figure of 4% [21]. These comparisons indicate that PSU may be more prevalent among midwifery populations compared to allied professions. Our study sample of 623 NMC registered midwives met the criteria for estimating a simple proportion of PSU within the population. However, with a greater proportion of younger midwives responding to the survey than in the midwifery population and methodological decisions around limiting demographic questions to protect participants’ identities, our ability to ensure full representativeness in our sample is inhibited. With the aim of the paper to address the lack of data on PSU in midwifery, we deemed threats to representativeness appropriate; however, results must be viewed with caution when attempting to identify prevalence.

Midwives within this sample reported substance use in response to work-related stress and anxiety, bullying, traumatic clinical incidents and maintenance of overall functioning. These findings are comparable with those in relation to paramedics, for whom PSU may also be linked to occupational and post-traumatic stress [22]. While 11% of this sample indicated they had sought help for PSU, 27% indicated that they did not seek help despite feeling they should. Within the responses, reported barriers to help-seeking included fear of repercussions, shame, stigma, practicalities and a perceived lack of support available or required. A recent paper calls for the development of structural interventions to tackle bullying, work-related stress and burnout currently endemic in the sector [10], and the COVID-19 pandemic that escalated after these data were collected may further psychologically deplete frontline healthcare workers [23]. Such action will also be required to support those engaged with PSU, particularly as such cases may now also be exacerbated by the COVID-19 pandemic [24]. Occupational health professionals have indicated that they do not feel adequately trained or resourced to support healthcare workers with PSU [25], and other gatekeepers to treatment may also be registered practitioners obliged to report impaired midwives to the NMC thus leaving midwives feeling unable to ask for support. Further training, education and attention in this area of occupational health for midwives, and allied health professionals, is required to ensure a compassionate approach.

Most perceptions in relation to impairment were compassionate within our sample. Yet, as evidenced by some of the illustrative quotes presented, punitive attitudes displayed towards impaired midwives can dissuade others from seeking help and thus prolong risks to both professionals and the public [11]. Broader naming, shaming and stigmatization of PSU as reported in our data may result in unjustified breaches of health-related confidentiality, as well as cause occupational, personal and professional harm, and should be challenged. Some acts reported here, such as the theft of drugs, may be behavioural symptoms of ill-health rather than cognizant acts contrary to probity due to the loss of behavioural control over drug-seeking and drug-taking, which has long defined quintessential addiction [26]. Thus, policies and actions which take a compassionate and non-punitive approach may be most useful in recovery and the development of safer working environments as supported elsewhere for allied professions [11,17].

The positive outcomes of the UK National Health Service Practitioner Health Programme (PHP) [27], established to support and treat doctors and other medical practitioners with PSU and addiction problems, as well as broader mental health problems, can provide a useful framework for the provision of a similar service for midwives. With doctors who had completed treatment though the PHP more likely to be in active employment at the end of the programme, a targeted employee assistance programme for midwives could help retain experienced and skilled midwives in the profession, as is needed to meet current workforce challenges [8]. The size of the population registered with the NMC may make replicating the PHP for this group problematic due to the scale of the resources required. However, using a theory-driven online intervention as proposed for work-related stress [28] may provide an efficient and effective solution to balancing scarce physical resources.

Our study is an exploratory first step to understanding PSU in a UK-based midwifery population and contributes to the existing evidence of substance use within the healthcare professions. Although the sample size is large, it only captures a small percentage of the UK midwifery population. This was expected due to the perceived risks associated with disclosure, alongside the decision to end data collection prematurely to prevent a distortion of results related to the COVID-19 pandemic. Additionally, the need to balance the risks associated with non-confidential disclosures compromises the generalization of these results and the self-selection of the participants’ risks bias through a desire to share strong opinions and experiences. Substance use is more common in younger age groups [29]; therefore, the greater proportion of younger respondents may lead to an over-estimation of the scale of PSU within our data. However, survey respondents typically under-report socially undesirable activities and over-report socially desirable ones [30]. This would indicate a potential for the under-reporting of PSU in our findings.

PSU in midwifery populations poses personal, professional and occupational risks of harm. It is not conducive to occupational well-being, nor excellence in the midwifery profession. As well as addressing the underlying causes of PSU with regards to this population, future research could usefully assess the prevalence of PSU, challenge stigmatizing perceptions with a view to increasing help-seeking behaviours and co-create effective bespoke interventions of support. Supporting occupational health providers, managers and policy makers to assist those engaged in PSU compassionately and with the assurance of confidentiality and anonymity may also augment help-seeking behaviours in this population, thus resulting in a reduction of risk and the maintenance of a skilled midwifery workforce.

Funding

No funding was provided for this project.

Acknowledgments

We would like to give thanks to the Royal College of Midwives, The British Journal of Midwifery and the Make Birth Better Network for sharing our recruitment advert within their networks, and to all the midwives who shared their experiences within the survey.

Competing interests

The lead author S.P. acts as a panellist for the Nursing and Midwifery Council’s Investigating Committee, presiding over cases brought before the council, including those related to problematic substance use. All other authors declare no conflicts of interest.