2019
Alcohol-related social and health
service use patterns as predictors of
death and remission in patients with AUD
Rautiainen, Elina
Elsevier BV
Tieteelliset aikakauslehtiartikkelit
© Elsevier Inc
CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/
http://dx.doi.org/10.1016/j.jsat.2018.10.013
https://erepo.uef.fi/handle/123456789/7350
Downloaded from University of Eastern Finland's eRepository
Alcohol-related social and health service use patterns as predictors of death and remission in patients with AUD
Elina Rautiainen, Olli-Pekka Ryynänen, Eeva Reissell, Jussi Kauhanen, Tiina Laatikainen
PII: S0740-5472(18)30241-1
DOI:
https://doi.org/10.1016/j.jsat.2018.10.013Reference: SAT 7786
To appear in:
Journal of Substance Abuse TreatmentReceived date: 25 May 2018
Revised date: 26 October 2018 Accepted date: 29 October 2018
Please cite this article as: Elina Rautiainen, Olli-Pekka Ryynänen, Eeva Reissell, Jussi Kauhanen, Tiina Laatikainen , Alcohol-related social and health service use patterns as predictors of death and remission in patients with AUD. Sat (2018),
https://doi.org/10.1016/j.jsat.2018.10.013
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.
ACCEPTED MANUSCRIPT
Alcohol-related social and health service use patterns as predictors of death and remission in patients with AUD
Elina Rautiainena, Olli-Pekka Ryynänena,b, Eeva Reissellc, Jussi Kauhanena , Tiina Laatikainena,c,d
a Institute of Public Health and Clinical Nutrition, PO Box 1627, FI-70211 University of Eastern Finland, Kuopio, Finland
b General Practice Unit, Kuopio University Hospital, Primary Health Care, PO Box 100, FI-70029 KUH, Kuopio, Finland
c National Institute for Health and Welfare, PO Box 30, FI-00271 Helsinki, Finland
d Joint Municipal Authority for North Karelia Social and Health Services (Siun sote), Tikkamäentie 16, 80210 Joensuu, Finland
*Corresponding author.
E-mail address: elinara@uef.fi (E. Rautiainen) Postal address:
University of Eastern Finland Kuopio campus
Institute of Public Health and Clinical Nutrition P.O. Box 1627
FI-70211 Kuopio, Finland
Abbreviations
AUD, alcohol use disorder; CI, confidence interval; EHR, electronic health record; HSU, health service utilization; MH, mental health; OR, odds ratio; PHC, primary health care; DST, Serum Desialotransferrin; GT, Plasma Glutamyl transferase; ALT, Plasma Alanine aminotransferase; AST, Plasma Aspartate aminotransferase; ALP, Plasma Alkaline phosphatase; MCV, mean corpuscular volume
ACCEPTED MANUSCRIPT
1. Introduction
Although patients with alcohol use disorders (AUDs) have been identified as one of the most expensive client groups with excess use of social and health care services (de Weert-van Oene, Termorshuizen, Buwalda, & Heerdink, 2017; Graham et al., 2017; Leskelä et al., 2013), the current knowledge of the patterns and continuity of alcohol-related use of social and health care services among this patient group is limited. Many studies have noted that substance use treatment services are heavily underutilized, and many people with AUDs are not receiving adequate care for their addiction. Previous studies have revealed that the treatment gap for AUDs is larger than in any other mental disorder (Roerecke & Rehm, 2014), and only approximately 1 in 7 of the patients with AUD use alcohol treatment services (Cohen, Feinn, Arias, &
Kranzler, 2007). Several researchers have identified barriers to AUD care, such as social stigma and unavailability of services (Grant, 1997; Keyes et al., 2010; Mojtabai, Chen, Kaufmann, & Crum, 2014;
Mojtabai, Olfson, & Mechanic, 2002; Saunders, Zygowicz, & D´Angelo, 2006). Male gender, being single, having low educational or income level, and having a mood or drug disorder, have been identified as predictors of receiving alcohol treatment (Cohen et al., 2007; Dawson, Goldstein, & Grant, 2012; Edlund, Booth, & Han, 2012; Twomey, Baldwin, Hopfe, & Cieza, 2015).
As many researchers examining alcohol-related health service utilization (HSU) have concentrated on specialized addiction treatment provision, there is little information available on the extent of alcohol- related HSU and continuity of care of patients with AUD across the whole social and health service delivery system. System approach concept highlights the importance of examining treatment services from a broader population health perspective beyond the provision of individual specialized services and single treatment episodes (Babor, Stenius & Romelsjö, 2008; Rush, 2010). Previous treatment system research has identified that individuals with AUD seek treatment from various sources and often from non-specialist services such as primary care physicians (Cohen et al., 2007; Dawson et al., 2012; Mowbray, Glass, &
Grinnell-Davis, 2015). This behavior is particularly true in the Finnish context, as the organization of primary health care (PHC), specialized mental care, and social services is fragmented, and the interoperability of patient information systems has been deficient. Notably, the co-occurring AUD and mental health (MH) problems have been associated with increased HSU and treatment episodes (Gonzalez & Rosenheck, 2002;
Graham et al., 2017; Kêdote, Brouselle & Champagne, 2008; Rush, Dennis, Scott, Castel, & Funk, 2008).
When examining the average annual use of primary care, emergency care, and hospitalizations, Graham et al. (2017) noted that patients with both MH and substance abuse problems were the most frequent users of all types of medical services compared with controls. Kêdote et al. (2008) noted in their prospective cohort study that the HSU was highest among patients with severe mental illness and co-occurring SUD.
Furthermore, a recent US-based study examined treatment outcomes associated with different alcohol-
ACCEPTED MANUSCRIPT
related service use patterns and, according to the results, alcoholics anonymous combined with specialty addiction services was associated with better treatment outcomes (Mowbray et al., 2015).
In Finland, alcohol-related HSU is traditionally estimated by using population surveys and the national health register (HILMO) data. However, the information on diagnoses of the PHC visits in the national social and health care registers are currently not comprehensive, leading to potential underestimation of the number of AUD-related visits (Kuussaari, Ronkainen, Partanen, Kaukonen, & Vorma, 2012). Furthermore, selection bias may exist in population surveys, causing erroneous results, as non-participants tend to have a higher risk of alcohol-related diseases and increased risk for hospitalization and death compared to participants (Gorman et al., 2014; Jousilahti, Salomaa, Kuulasmaa, Niemelä, & Vartiainen, 2005; Karvanen, Tolonen, Härkönen, Jousilahti, & Kuulasmaa, 2016).
This register study aims to examine the extent and continuity of alcohol-related service utilization of patients with AUD across the social and health service system, including various social and health professional groups, and to assess whether the alcohol-related social and health service use predict different outcomes, such as death or remission. By using linked primary and secondary care electronic health records (EHRs) and municipal social services client databases, it is possible to thoroughly examine the alcohol-related service utilization and factors associated with outcomes of death and remission.
Municipal social services client databases have not been extensively utilized in register-based alcohol- related HSU research before this study.
2. Materials and methods
2.1. Data source and treatment system
Data on alcohol-related social and health service use was gathered from the EHRs and municipal social services client databases in the North Karelia region of Eastern Finland, for the years 2011-2016. The study was approved by the Research Ethics Committee of the Kuopio University Hospital. Consent was not obtained, as the study was based on registry information.
North Karelia is one of the first areas in Finland to have prepared for the impending social and health care reform, by establishing a structured, integrated EHR system. This integrated EHR system is used across municipalities in primary and specialized care, as well as in adult social work and specialized AUD services (specialized AUD clinic), enabling a thorough analysis of HSU, including all visits to health and social care professionals, laboratory measures and treatment episodes. Data on income support was retrieved from the municipal social services client database.
North Karelia has approximately 165,000 inhabitants. The alcohol-related mortality in the region has been above the national average. In Finland, AUD treatment is organized and provided by health services,
ACCEPTED MANUSCRIPT
specialized addiction treatment services, and in social care. The role of primary care is emphasized in AUD detection. Currently, the organization of PHC, and specialized mental care and social services is fragmented, causing challenges to the care-coordination (Finnish Medical Society Duodecim, 2015).
2.2. Study sample
The study sample was formed retrospectively, based on the medical diagnoses (WHO, 2016) in the EHR register. A total of 3935 working-aged (18–65 years) individuals had an alcohol-related visit to health services between the years 2011–2012, from which a random cohort (n=396) was formed retrospectively and followed prospectively in time, from January 2011 until December 2016. Alcohol-related visits included the following ICD-10 codes: F100, F101, F102, F103, F104, F105, F106, F108, F109, G312, G405, G4050, G4051, G4052, G621, I426, K292, K860, K700, K701, K702, K703, K704, K709, T510, T511, T512, T513, T518, T519, X45, and X69.
2.3. Measures
2.3.1. Outcome variables
For the HSU comparison, the data were divided into three mutually exclusive categories, according to the outcome status at the end of the follow-up period: 1) dead, 2) present AUD, and 3) AUD in remission.
The date of death was automatically linked to the EHR through the population register center. Present AUD was defined as having alcohol-related visits (ICD-10 code F100, F101, F102, F103, F104, F105, F106, F108, or F109, as the main diagnosis) and notifications of harmful use of alcohol or dependence, in the EHR notes in each year of the follow-up period. Patients were classified in the present AUD group also if they had alternating abstinence periods (max. few months) followed by relapses. Remission was defined as sustained abstinence or managed use that lasted until the end of the follow-up period with minimum duration of six months. Assessment of the time estimate in AUD remission was based on health professionals´ objective notes and diagnosis information, i.e., the notes systematically identifying the patient as abstinent or managing their alcohol use, or the patient had ICD-10 diagnosis codes F1020–F1023, indicating sustained remission. In case of mixed reviews between the health professionals’ notes, the patient was classified as having present AUD. The cohort was identified as having advanced form of AUD and the annual proportion of those achieving sustained remission varied between 3-5%. (Rautiainen et al., 2018). Patients with no comments on AUD status due to lack of yearly visits were excluded. Definition of alcohol use disorder based on the Diagnostic and Statistical Manual of Mental Disorders (DSM‐IV) and International Statistical
ACCEPTED MANUSCRIPT
Classification of Disease (ICD‐10) and included harmful use and alcohol dependence. All study subjects filled this definition at the baseline.
2.3.2. Socioeconomic characteristics
Socioeconomic variables included age, gender, marital status, unemployment status, homelessness, illicit drug use, criminal record, and drunk driving. Information regarding socioeconomic factors was manually collected from the EHR notes. Income support data was obtained from the municipal social services client databases. Unemployment status, homelessness, illicit drug use, criminal background, and drunk driving variables were classified as two-class variables, according to whether the study participant had any such mentions in the EHR within the 6-year period, or not. Drunk driving was recorded in the EHR notes when police authority requested a medical expert’s opinion (i.e. drivers´ license monitoring period with duration up to 3-6 months, including clinical examination and laboratory measures) on substance abuse status in case of identified drunk driving event.
2.3.3. Clinical variables
Information regarding comorbidity (i.e., permanent ICD-10 codes) was collected from the routinely compiled EHR statistics. In the EHR, permanent diagnosis is used for chronic or long-term diseases that are considered to affect the extended care of the patient. These diagnoses remain in the EHR even after the disease is cured. Permanent alcohol diagnosis was defined as ICD-10 codes F100, F101, F102, F103, F104, F105, F106, F108, or F109 (mental and behavioral disorders due to alcohol use), and permanent MH diagnosis as ICD-10 codes F00–F99 (mental and behavioral disorders), excluding F10 codes, respectively.
Permanent diagnoses were classified into three groups, according to number; 1) none, 2) one, and 3) two or more (indicating multimorbidity). Laboratory measures, which were used for identification, treatment, and assessment of care of AUD, included serum desialotransferrin (S-DST), plasma glutamyl transferase (P- GT), plasma alanine aminotransferase (P-ALT), plasma aspartate aminotransferase (P-AST), plasma alkaline phosphatase (P-ALP), and mean corpuscular volume (E-MCV), which were calculated as a yearly mean number of measures, by considering the eligibility period of the study subjects (i.e. the number of laboratory measures of the 6-year follow-up was divided by the eligibility time).
2.3.4. HSU variables
HSU comprised 1) primary care services, including alcohol-related (F10) doctor visits, other MH-related doctor visits, visits to psychologists, and alcohol-related inpatient treatment episodes in a primary care
ACCEPTED MANUSCRIPT
ward, 2) specialized AUD services, including doctor visits, nurse visits, social worker visits, detoxification treatment, rehabilitation, evaluation periods, housing rehabilitation, interval treatment, crisis treatment, and sobriety support, 3) primary care level MH services, (i.e., MH units that operate as part of PHC services), including mental health-related doctor visits and nurse visits, and 4) specialized care visits, including alcohol-related (F10) doctor visits and alcohol-related inpatient treatment episodes, as well as doctor, nurse, psychologist, and social worker visits for psychiatric treatment.
Alcohol-related primary and specialized health care service utilization was defined as having one of the following ICD-10 codes as the main diagnosis for the visit: F100, F101, F102, F103, F104, F105, F106, F108, or F109. Somatic damage caused by alcohol, such as accidents and other somatic disorders, were excluded from this analysis and will be examined in further studies. Primary care doctor visits for MH reasons were defined as having an ICD-10 code F01–F99 (excluding F10 codes), as the main diagnosis for the visit.
Information regarding institutionalization (i.e. long-term inpatient treatment, sheltered housing etc.) was collected from the EHR.
Temporal relationship between HSU and outcomes was assessed by using eligibility period as a method to adjust for the number of years of observation for a given patient. That is, alcohol-related HSU measures were constructed by converting the total number of visits to yearly means by dividing the total number of visits with the eligibility period.
2.4. Statistical analysis
IBM SPSS Modeler (version 18.0, SPSS Inc, Chicago, IL, USA) was used to derive the health service use variables from the EHR data, and IBM SPSS Statistics 24 (SPSS Inc, Chicago, IL, USA) was used in the statistical analyses. The X2, Fischer’s exact and Kruskal–Wallis tests were used for the group comparisons, and bivariate and multivariate Cox-regression analyses were used to further examine factors predicting death and remission. First, all the variables were tested separately in a bivariate model, and those with a significance level of p ≤0.05 were further included in a multivariate model. Then, by using backward elimination, statistically insignificant variables were excluded from the final multivariate models.
Proportional hazards assumption was evaluated using log-log-plots. Multivariate analysis, with death as the outcome variable, were adjusted for marital status, number of permanent diagnoses, drunk driving, doctor visits to specialized AUD clinic, social worker visits, doctor visits to MH services, and alcohol-related visits to PHC doctor, i.e., only variables with a significance level of p ≤0.05 remained in the model. Multivariate analysis with remission as the outcome variable was adjusted for institutional care, nurse visits to specialized AUD clinic, rehabilitation at specialized AUD clinic, MH nurse visits, alcohol-related visits to PHC doctor, other MH-related visits to PHC doctor, alcohol-related doctor visits to specialized AUD clinic, and the number of DST laboratory measures. The results of the Cox-regression analyses are presented with
ACCEPTED MANUSCRIPT
odds ratios (OR) and 95% confidence intervals (CI) limits. The significance level was set at p ≤0.05, p≤0.01, or p ≤0.001.
3. Results
3.1. Baseline characteristics
Table 1 provides the baseline characteristics of the cohort, according to the outcome status. The majority of the study subjects received income support during the follow-up period, indicating a low socio- economic status. In addition, a total of 48% had experienced at least one unemployment period during the follow-up period. Mentions of homelessness were gathered from the EHR notes, but this was not frequently recorded information; only 6.8% of the study subjects had any mentions regarding homeless status. A total of 17.4% of the study subjects were illicit drug users, and illicit drug use was most prevalent among those with present AUD. Criminal background (i.e., mentions of prison sentence) was recorded for 19.4% of the study subjects, and drunk driving for 18.7% of the study subjects, respectively. Figure 1 presents the number of chronic disorders by age group.
3.2. Alcohol-related and MH service use
Table 2 presents the frequencies of alcohol-related and MH-related HSU, according to different service providers and social and health professional groups, and according to the outcome group. PHC was the most frequently used service, and the majority of the study subjects had at least one directly alcohol- related visit to a primary care doctor, during the follow-up period. However, significant differences between the outcome groups existed. For instance, 63% of those achieving remission had no alcohol- related primary care doctor visits or alcohol-related hospitalizations during the follow-up period. Instead, these subjects had comparatively more frequent use of specialized AUD services, as well as an increased number of visits to primary care doctor for other MH reasons. Respectively, the vast majority of those who died had no MH-related visits to primary care doctor or presented as MH service users.
Visits to a doctor in specialized AUD clinic were infrequent, and most of the study participants did not have access to a doctor in specialized AUD services. In comparison, visits to a nurse in specialized AUD services were more common, although significant differences between the outcome groups existed, and the majority (60.2%) of those who died had no contact with nurses in the specialized AUD services.
Furthermore, specialized inpatient AUD treatment was not extensively utilized.
Alcohol-related visits to specialized health care (i.e., emergency care visits) included visits to all the specialties, with ICD-10 code F10 as the main diagnosis. Approximately half of the study subjects (52%) had
ACCEPTED MANUSCRIPT
at least one such visit, although outcome groups differed significantly, and most (72%) of those achieving remission had no alcohol-related specialized health care visits. The same trend was noted, when examining alcohol-related hospitalizations (inpatient care episodes) in specialized health care, with alcohol-related diagnosis as the main diagnosis. Also, specialized care psychiatric specialty visits were reviewed, and those who died had significantly fewer visits to a psychiatrist or psychiatric nurse, compared with the other outcome groups. Table 3 lists the frequencies of laboratory measures, which were used for treatment motivation, the identification of AUDs, and assessment of AUD care.
3.3. Clinical- and health service use-related variables as predictors of remission
Table 4 summarizes the results of the Cox-regression analyses, examining the association between socioeconomic-, clinical-, and health service use-related factors and remission. In the multivariate model, institutional care significantly increased the odds of remission (OR 1.55, p<0.001, 95% CI: 1.33–1.81), as did the contact with a MH nurse (OR 1.22, p=0.034, 95% CI: 1.02–1.46) and visits to the PHC doctor for MH reasons (ICD-10 codes F00–F99, excluding F10) (OR 1.55, p=0.007, 95% CI: 1.13–2.13). Rehabilitation in the specialized AUD clinic was not statistically significant in the bivariate model, but in the multivariate model (model 2), it was associated with increased odds of remission (OR 6.75, p<0.001, 95% CI: 2.27–20.11).
Furthermore, a more frequently measured serum DST value increased the odds of achieving remission (OR 1.27, p< 0.001, 95% CI: 1.16–1.39). Alcohol-related (F10 diagnosis) visits to a PHC doctor were associated with decreased odds of achieving remission (OR 0.65, p=0.016, 95% CI: 0.46–0.92), as were the alcohol- related visits to specialized care doctor (OR 0.54, p< 0.005, 95% CI: 0.35–0.83).
3.4. Socioeconomic-, clinical-, and health service use-related variables as predictors of death
Table 5 shows the results of the Cox-regression analyses, examining whether socioeconomic, clinical, and health service use variables were associated with the risk of death. The bivariate analysis (model 1) presents all the statistically significant variables (p<0.05, 95% CI), following the multivariate analysis with adjustments (model 2). The number of permanent diagnoses was related to decreased odds of death, particularly in the cases with multimorbidity (2+ permanent diagnoses) (OR 0.19, p<0.001, 95% CI: 0.11–
0.33). Also, drunk driving (OR 0.28, p=0.002, 95% CI: 0.13–0.62) and visits to doctor in specialized AUD clinic decreased the odds of death (OR 0.36, p<0.001, 95% CI: 0.21–0.60), as did the doctor visits to MH clinics (OR 0.50, p=0.036, 95% CI: 0.26–0.96). Instead, alcohol-related visits to a primary care doctor (F10 diagnosis) were associated with increased odds of death (OR 1.67, p<0.001, 95% CI: 1.28–2.18).
ACCEPTED MANUSCRIPT
4. Discussion
This study examined the relationship of alcohol-related social and health service utilization frequency to long-term care outcomes of remission and death in patients with AUD. We applied a treatment system approach (Babor, Stenius & Romelsjö, 2008) and examined the wide variety of alcohol-related service use, including services that lie outside the formal specialized AUD service provision that may have relevance in the treatment career over the individual´s life course. By comparing the study participants’ HSU according to the outcome status, we identified frequent use of mental health services provided through primary care to be associated with better treatment outcomes. Conversely, we identified those who died as the most frequent visitors to primary care doctor for alcohol-related reasons and as the most infrequent users of the specialized AUD services. In the multivariate analysis the frequent alcohol-related visits to PHC doctor were associated with increased odds of death (OR 1,67) but could not be explained by the multimorbidity (i.e.
that the persons with excess PHC doctor visits were sicker); instead, an increased number of permanent diagnoses was associated with decreased odds of death, most likely indicating better treatment contact and thereby better detection of underlying diseases, and more comprehensive coding of permanent diagnoses.
The outcome status largely reflected the continuity and quality of care received (i.e. treatment career) of individuals with an advanced state of AUD, as the EHRs-based random sampling captured mostly individuals with a disadvantaged background and long history of AUD, which was noted in our previous research (Rautiainen et al., 2018). Thereby, the use of EHR data enabled us to examine the alcohol-related HSU of the most deprived population, who would be most likely to drop out from the population survey studies (Gorman et al., 2014). Thus, this EHR-based register analysis of the alcohol-related social and HSU of primary and specialized care, and the use of MH and social care services for 6 years provided improved information on alcohol-related HSU by socioeconomically deprived individuals with advanced AUD, in relation to long-term care outcomes. The findings of our study revealed significant differences in the frequencies of HSU use and differences in access to social and health professionals among the outcome groups.
Detailed examination of the extent of alcohol-related HSU showed that this cohort had substantially better access to a PHC doctor for alcohol-related reasons than to a doctor in the specialized AUD services.
Also, inpatient detoxifications in the PHC ward were more common than inpatient detoxification at the specialized AUD services. This result reflects the current organization of AUD treatment, where the service delivery emphasis is on the primary care, but it may also reflect inadequate resources in specialized AUD services and difficulties in accessing these services. Our results from the multivariate analyses indicated that there are challenges in the AUD treatment in primary care settings, as frequent alcohol-related visits to a primary care doctor were associated, not with an improved outcome, but with a reduced probability of
ACCEPTED MANUSCRIPT
achieving remission and even with the increased risk of death. Alternatively, access to a doctor in specialized AUD clinics was associated with a decreased risk of death during follow-up (OR 0.36).
Mental health problems are known to be highly prevalent among individuals with AUD (Grant et al., 2004; Kenneson, Funderburk & Maisto, 2013). As AUDs may also be treated in MH settings, we included the use of MH services in our analyses. We noted that an increase in the frequency of visits to MH service nurses was associated with increased odds of remission (OR 1.22) and the contact with MH service doctors was associated with a decreased risk of death (OR 0.50). Also, more frequent contact with a PHC doctor for MH reasons (other than F10 dg) was associated with increased odds of remission (OR 1.89). Other factors associated with an increased probability of achieving remission included institutionalization, frequent measurement of the laboratory biomarker serum DST, and rehabilitation in specialized AUD clinic. These results support the previous studies, identifying that patients treated in specialized addiction clinics or psychiatric specialty care received more comprehensive care and had better treatment outcomes compared with those treated only in primary care (Moos, Finney, Federman, & Suchinsky, 2000; Mowbray et al., 2015; Ray, Weisner, & Mertens, 2005). McLellan et al. (1996) showed that a broad variety of services and high frequent service use led to better addiction treatment outcomes. The findings by Grella, Stein, Weisner, Chi, and Moos (2010) suggested that continuity of care was associated with decreased substance use disorder and MH symptoms and Proctor & Herschman (2014) concluded that the continuity of care is essential in achieving successful long-term outcomes.
We identified nurses as the most frequently visited professional group accessed through MH and specialized AUD services. The results from the multivariate analyses showed that frequent nurse visits in MH settings were associated with better treatment outcomes (i.e., achieving remission), whereas visits to nurses in specialized AUD services did not have a similar effect. Rehm et al. (2015) identified that those accessing specialized AUD services typically have severe form of AUD. Conversely, high psychiatric comorbidity has also been recognized as a barrier to treatment effectiveness (Weisner, Matzger, &
Kaskutas, 2003). Our study cohort had an advanced AUD and mental health co-morbidity was most prevalent among those achieving stable remission. Most likely MH problems were better identified in the remission group and these patients received more comprehensive care and were better committed to care compared to other outcome groups. In future, it would be interesting to examine the potential effect of training level of nurses (i.e., assistant nurse, registered nurse, specialized nurse or advanced nurse practitioner) on quality of care and AUD care outcomes.
Lastly, our study suggests that the number of permanent diagnoses and frequency of testing the serum DST laboratory measure differed notably among the outcome groups and were lowest among those who died, despite their reasonably frequent contact with PHC doctor. This trend may either reflect the lack of perceived need for treatment, or low commitment to treatment by the study participants. However, these findings may also indicate undiagnosed and untreated illnesses, i.e., challenges in care quality and in
ACCEPTED MANUSCRIPT
the ability to engage patients in the care. Also, the observed relatively high amount of emergency care visits among those who died supports this interpretation of unmet care needs. Integrated MH and AUD care at the primary care level would most likely enhance the care outcomes among this patient group, as has been suggested earlier (Kessler et al., 1996; Lundgren et al., 2014). The integrated EHRs are an essential tool in this integration process, as Tai, Wu, and Clark (2012) have pointed out.
4.1. Limitations
This study had several limitations that should be addressed. First, there are limitations in the use of EHRs in outcome research, as these records are not designed primarily for research purposes. Especially, the chronic and relapsing nature of AUDs can be difficult to capture. In this study, the challenge of alternating AUD status was resolved by considering the individuals with such alternating periods as having current AUD and outlining remission to those achieving stable remission that lasted at least 6 months and until the end of the follow-up period. Naturally, it is possible that some of those achieving stable remission may have relapsed after the follow-up period ended. Second, with regard to EHR data quality, a total of 1435 visits to a PHC doctor had a missing diagnosis. The proportion of alcohol-related missing diagnoses was 23.7%. This gap imposes challenges regarding the register data quality and may lead to underestimation of the prevalence of AUDs. The present study addressed this issue by manually reviewing all the primary care EHR records with missing diagnoses. It should be also noted that MH services may have included alcohol-related mental health visits, as it is possible that MH services have been used to compensate the under-resourced specialized AUD services. Third, socio-economic status (SES) variables most likely have changed during the course of the study period, which the binary coding was not able to capture. Binary coding with most recent information was used as SES was inconsistently recorded in the EHR. Fourth, as the EHRs-based random sampling captured mostly individuals with a disadvantaged background, the results are not generalizable to the whole working-age population. Likewise, the results of this study are fully representative only in the North Karelia district in Finland, as the organization of AUD treatment may vary globally and even within the country and between municipalities. Lastly, this study was not able to assess private health service use or occupational health service use provided by private providers, as they use different EHR systems. However, study participants receiving income support are highly unlikely to use private services, and private institutes that were offering occupational health services existed in two municipalities only.
ACCEPTED MANUSCRIPT
5. Conclusions
This EHR-based register analysis provided a detailed examination of alcohol-related social and health services use among socioeconomically deprived individuals with an advanced form of AUD and investigated alcohol-related HSU patterns in relation to AUD care outcomes. Significant differences in the frequency of HSU were observed among the three outcome groups, who either achieved remission, remained with the AUD, or died during follow-up. Our study identified the importance of MH services provided through primary care, as frequent use of these services were associated with enhanced AUD treatment outcomes.
Conversely, use of PHC services alone was insufficient for AUD patients with deprived background, as frequent alcohol-related visits to PHC doctor were associated with an increased risk of death. Thus, patients treated also in specialized AUD services or MH services had better treatment outcomes compared with those treated mostly in primary care. As treatment of AUD is currently organized mainly in primary health care, our results supported the integration of MH and specialized AUD services into PHC services, to improve treatment outcomes and access to care. Also, indicators for measuring the quality of care among patients with AUD should be discussed.
Funding
Elina Rautiainen was supported by the Finnish Foundation for Alcohol Studies, and the University of Eastern Finland Graduate School. This study received funding from the Strategic Research Council at the Academy of Finland (312703, 312708). The funding sources had no further role in the study design, including the collection, analysis, and interpretation of data, the writing of the report, or in the decision to submit the paper for publication.
Contributors
T. Laatikainen, O-P. Ryynänen, and E. Rautiainen participated in planning and designing the study. E.
Rautiainen performed all data management and analysis and drafted the article. T. Laatikainen, O-P.
Ryynänen, J. Kauhanen, and E. Reissell critically reviewed the document. All authors contributed to and have approved the final manuscript.
Conflict of interest: none.
Acknowledgments
We wish to thank research assistant Laura Kekäläinen for the valuable help with the data collection.
ACCEPTED MANUSCRIPT
References
Babor, T.F., Stenius, K. & Romelsjö, A. (2008). Alcohol and drug treatment systems in public health perspective: mediators and moderators of population effects. Int Journal of Methods in Psychiatric Research, 17(S1), S50-S59.
Cohen, E., Feinn, R., Arias, A., & Kranzler, H. R. (2007). Alcohol treatment utilization: Findings from the National Epidemiological Survey on Alcohol and Related Conditions. Drug and Alcohol Dependence, 86(2–3), 214–221.
Dawson, D. A., Goldstein, R. B., & Grant, B. F. (2012). Factors associated with first utilization of different types of care for alcohol problems. Journal of Studies on Alcohol and Drugs, 73(4), 647–656.
de Weert-van Oene, G. H., Termorshuizen, F., Buwalda, V. J. A., & Heerdink, E. R. (2017). Somatic health care utilization by patients treated for substance use disorders. Drug and Alcohol Dependence, 178, 277–
284.
Edlund, M. J., Booth, B. M., & Han, X. (2012). Who seeks care where? Utilisation of mental health and substance use disorder treatment in two national samples of individuals with alcohol use disorders. Journal of Studies on Alcohol and Drugs, 73(4), 635–646.
Finnish Medical Society Duodecim. Current Care Guidelines. Treatment of alcohol abuse. (2015). Working group appointed by the Finnish Medical Society Duodecim and the Finnish Society of Addiction Medicine,
Helsinki (published November 9, 2015).
http://www.kaypahoito.fi/web/english/guidelineabstracts/guideline?id=ccs00005 Accessed 15 March 2018.
Gonzalez, G., & Rosenheck, R. A. (2002). Outcomes and service use among homeless persons with serious mental illness and substance abuse. Psychiatric Services, 53(4), 437–446.
Gorman, E., Leyland, A. H., McCartney, G., White, I. R., Katikireddi, S. V., Rutherford, L., . . . Gray, L. (2014).
Assessing the representativeness of population-sampled health surveys through linkage to administrative data on alcohol-related outcomes. American Journal of Epidemiology, 180(9), 941–948.
Graham, K. G., Cheng, J., Bernards, S., Wells, S., Rehm, J., & Kurdyak, P. (2017). How much do mental health and substance use/addiction affect use of general medical services? Extent of use, reason for use, and associated costs. Canadian Journal of Psychiatry, 62(1), 48–56.
Grant, B. F. (1997). Barriers to alcoholism treatment: Reasons for not seeking treatment in a general population sample. Journal of Studies on Alcohol and Drugs, 58(4), 365–371.
Grant, B.F., Stinson, F.S., Dawson, D.A., Chou, S.P., Dufour, M.C., Compton, W., et al. Kaplan, K. (2004).
Prevalence and co-occurrence of substance use disorders and independent mood and anxiety disorders:
results from the National Epidemiologic Survey on Alcohol and Related Conditions. Archives of General Psychiatry, 61, 807–816. doi: 10.1001/archpsyc.61.8.807.
Grella, C. E., Stein, J. A., Weisner, C., Chi, F., & Moos, R. (2010). Predictors of longitudinal substance use and mental health outcomes for patients in two integrated service delivery systems. Drug and Alcohol Dependence, 110(1–2), 92–100.
ACCEPTED MANUSCRIPT
Jousilahti, P., Salomaa, V., Kuulasmaa, K., Niemelä, M., & Vartiainen, E. (2005). Total and cause specific mortality among participants and non-participants of population based health surveys: A comprehensive follow up of 54 372 Finnish men and women. Journal of Epidemiology & Community Health, 59(4), 310–315.
Karvanen, J., Tolonen, H., Härkönen, T., Jousilahti, P., & Kuulasmaa, K. (2016). Selection bias was reduced by recontacting nonparticipants. Journal of Clinical Epidemiology, 76, 209–217.
Kêdote, M.N., Brousselle, A. & Champagne, F. (2008). Use of health care services by patients with co- occurring severe mental illness and substance use disorders. Mental Health and Substance Use, 1, 216-227.
Kenneson, A., Funderburk, J.S., Maisto, S.A. (2013). Substance use disorders increase the odds of subsequent mood disorders. Drug and Alcohol Dependence, 133, 338–343. doi:
10.1016/j.drugalcdep.2013.06.011.
Kessler, R. C., Nelson, C. B., McGonagle, K. A., Edlund, M. J., Frank, R. G., & Leaf, P. J. (1996). The epidemiology of co-occurring addictive and mental disorders: Implications for prevention and service utilization. American Journal of Orthopsychiatry, 66(1), 17–31.
Keyes, K. M., Hatzenbuehler, M. L., McLaughlin, K. A., Link, B., Olfson, M., Grant, B. F., & Hasin, D. (2010).
Stigma and treatment for alcohol disorders in the United States. American Journal of Epidemiology, 172(12), 1364–1372.
Kuussaari, K., Ronkainen, J. E., Partanen, A., Kaukonen, O., & Vorma, H. (2012). Kuntoutuksesta katkaisuun.
Päihdetapaslaksenta palvelujärjestelmän kuvaajana [The 2011 survey of intoxicant-related cases in Finland: a service system perspective]. Yhteiskuntapolitiikka [Journal of Social and Public Policies], 77(6), 640–655.
Leskelä, R.-L., Komssi, V., Sandström, S., Pikkujämsä, S., Haverinen, A., Olli, S.-L., & Ylitalo-Katajisto, K.
(2013). Paljon sosiaali-ja terveyspalveluja käyttävät asukkaat Oulussa. [Heavy users of social and health care services in the city of Oulu] Suomen Lääkärilehti [Finnish Medical Journal], 68, 3163–3169.
Lundgren, L., Wilkey, C., Chassler, D., Sandlund, M., Armelius, B-Å., Armelius, K., & Brännström, J. (2014).
Integrating addiction and mental health treatment within a national addiction treatment system: Using multiple statistical methods to analyze client and interviewer assessment of co-occurring mental health problems. Nordic Studies on Alcohol and Drugs, 31(1), 59–79.
McLellan, A. T., Woody, G. E., Metzger, D., McKay, J., Durell, J., Alterman, A. I., & O´Brien, C. P. (1996).
Evaluating the effectiveness of addiction treatments: Reasonable expectations, appropriate comparisons.
Milbank Quarterly, 74(1), 51–85.
Mojtabai, R., Chen, L.-Y., Kaufmann, C. N., & Crum, R. M. (2014). Comparing barriers to mental health treatment and substance use disorder treatment among individuals with comorbid major depression and substance use disorders. Journal of Substance Abuse Treatment, 46(2), 268–273.
Mojtabai, R., Olfson, M., & Mechanic D. (2002). Perceived need and help-seeking in adults with mood, anxiety, or substance use disorders. Archives of General Psychiatry, 59(1), 77–84.
Moos, R. H., Finney, J. W., Federman, E. B., & Suchinsky, R. (2000). Specialty mental health care improves patient´s outcomes: Findings from a nationwide program to monitor the quality of care for patients with substance use disorders. Journal of Studies on Alcohol, 61(5), 704–713.
ACCEPTED MANUSCRIPT
Mowbray, O., Glass, J. E., & Grinnell-Davis, C. L. (2015). Latent class analysis of alcohol treatment utilization patterns and 3-year alcohol related outcomes. Journal of Substance Abuse Treatment, 54, 21–28.
Proctor, S. L., & Herschman, P. L. (2014). The continuing care model of substance use treatment: what works, and when is “enough,” “enough?” Psychiatry Journal, 2014, 1–16.
http://dx.doi.org/10.1155/2014/692423.
Ray, G. T., Weisner, C. M., & Mertens, J. R. (2005). Relationship between use of psychiatric services and five-year alcohol and drug treatment outcomes. Psychiatric Services, 56(2), 164–171.
Rehm, J., Allamani, A., Aubin, H-J., Della Vedova, R., Elekes, Z., Frick, U., . . . Wojnar, M. (2015). People with alcohol use disorders in specialized care in eight different European countries. Alcohol and Alcoholism, 50(3), 310–318.
Roerecke, M., & Rehm, J. (2014). Cause-specific mortality risk in alcohol use disorder treatment patients: A systematic review and meta-analysis. International Journal of Epidemiology, 43(3), 906–919.
Rush, B. (2010). Tiered frameworks for planning substance use service delivery systems: Origins and key principles. Nordic Studies on Alcohol and Drugs, 27, 617-635.
Rush, B. R., Dennis, M. L., Scott, C. K., Castel, S., & Funk, R. R. (2008). The interaction of co-occurring mental disorders and recovery management checkups on substance abuse treatment participation and recovery.
Evaluation Review, 32(1), 7–38.
Saunders, S. M., Zygowicz, K. M., & D´Angelo, B. R. (2006). Person-related and treatment-related barriers to alcohol treatment. Journal of Substance Abuse Treatment, 30(3), 261–270.
Tai, B., Wu, L-T., & Clark, H. W. (2012). Electronic health records: Essential tools in integrating substance abuse treatment with primary care. Substance Abuse and Rehabilitation, 3, 1–8.
Twomey, C. D., Baldwin, D. S., Hopfe, M., & Cieza, A. (2015). A systematic review of the predictors of health service utilisation by adults with mental disorders in the UK. BMJ Open, 5(7), e007575.
http://dx.doi.org/10.1136/bmjopen-2015-007575
Weisner, C., Matzger, H., & Kaskutas, L. A. (2003). How important is treatment? One-year outcomes of treated and untreated alcohol-dependent individuals. Addiction, 98(7), 901–911.
WHO (2016). Classification of Diseases (ICD). International Statistical Classification of Diseases and Related Health Problems 10th Revision. http://apps.who.int/classifications/icd10/browse/2016/en
ACCEPTED MANUSCRIPT
Table 1. Patient characteristics by outcome group
Outcome 2016 Present AUD (n=228)
Dead (n=93)
Remission
(n=75) Total (n=396)
n % n % n % n % p
Kruskall- Wallis (CI95%)
Age at baseline 0.57a <0.001 p-d
18-24 12 5.3 1 1.1 3 4.0 16 4.0
25-34 28 12.3 7 7.5 9 12.0 44 11.1
35-44 45 19.7 12 12.9 12 16.0 69 17.4
45-54 89 39.0 33 35.5 26 34.7 148 37.4
55-64 54 23.7 40 43 25 33.3 119 30.1
Gender
Male 165 72.4 75 80.6 55 73.3 295 74.5 0.294a
Female 63 27.6 18 19.4 20 26.7 101 25.5
Marital status (n=332)
Single 76 39.8 44 58.7 20 30.3 140 42.2 0.009a 0.006 d-r
Married/cohabitation 54 28.3 17 22.7 24 36.4 97 29.2
Divorced/widow 61 31.9 14 18.7 22 33.3 95 28.6
Multimorbidity <0.001a <0.001 p-d
0 51 22.4 38 40.9 11 14.7 100 25.3 <0.001 d-r
1 41 18.0 19 20.4 18 24.0 78 19.7
2+ 136 59.6 36 38.7 46 61.3 218 55.1
Permanent Dg F10 0.001a <0.001 p-r
Yes 89 39.0 26 28.0 13 17.3 128 32.3
No 139 61.0 67 72.0 62 82.7 268 67.7
Permanent Dg F 0.019a 0.048 p-d
Yes 55 24.1 6 11.8 21 28.0 87 22.0 0.036 d-r
No 173 75.9 10 88.2 54 72.0 309 78.0
Institutional care
Yes 26 11.4 5 5.4 15 20 46 11.6 0.013a
No 202 88.6 88 94.6 60 80 350 88.4
Income support (N=382)
Yes 163 75.5 59 64.1 48 64.9 112 29.3 0.068a
No 53 24.5 33 35.9 26 35.1 270 70.7
Homeless
Yes 15 6.6 9 9.7 3 4.0 27 6.8 0.340a
No 213 93.4 84 90.3 72 96.0 369 93.2
ACCEPTED MANUSCRIPT
Unemployment
period 0.47a 0.040 p-d
Yes 118 51.8 31 33.3 41 54.7 190 48.1
No 63 27.6 34 36.6 27 36.0 124 31.3
Missing info 47 47.0 28 30.1 7 9.3 82 20.7
Relative died 0.464a
Yes 24 10.5 5 5.4 8 10.7 37 9.3
No 187 82.0 73 78.5 66 88.0 326 82.3
Missing info 17 7.5 15 16.1 1 1.3 33 8.3
Illicit drug use 0.073a
Yes 48 21.1 13 14.0 8 10.7 69 17.4
No 180 78.9 80 86.0 67 89.3 327 82.6
Criminal background 0.509a
Yes 47 20.6 19 20.4 11 14.7 77 19.4
No 181 79.4 74 79.6 64 85.3 319 80.6
Drunk driving 0.009a 0.007 p-d
Yes 53 23.2 8 8.6 13 17.3 74 18.7
No 175 76.8 85 91.4 62 82.7 322 81.3
a Pearson chi-square *p=present AUD, d=dead, r=remission
b Fischer's exact test
F10= ICD-10 codes F10.0-F10.9 (i.e. alcohol-related visit)
F= ICD-10 codes F00–F99 (mental and behavioral disorders), excluding F10 codes
ACCEPTED MANUSCRIPT
Figure 1. Number of chronic disorders by age group
0% 20% 40% 60% 80% 100%
18-24 25-34 35-44 45-54 55-64
5+
4 3 2 1 0
ACCEPTED MANUSCRIPT
Table 2. Yearly mean number of alcohol-related visits and other mental health-related visits to health care services, by outcome group
Outcome 2016 Present AUD (n=228)
Dead (n=93)
Remission (n=75)
Total (n=396)
n % n % n % n % p Kruskall-Wallis
(CI95%) Primary healthcare
Doctor visits (F10) <0.001a <0.001 p-r
0 55 24.1 30 32.3 47 62.7 132 33.3 <0.001 d-r
<1 118 51.8 31 33.3 15 20.0 164 41.4
1.0-1.9 43 18.9 16 17.2 7 9.3 66 16.7
2.0+ 12 5.3 16 17.2 6 8.0 34 8.6
Doctor visits (F) <0.001b <0.01 p-d
0 128 56.1 72 77.4 36 48.0 236 59.6 <0.001 d-r
<1 88 38.6 14 15.1 26 34.7 128 32.3
1.0-1.99 11 4.8 3 3.2 6 8.0 20 5.1
2.0+ 1 0.4 4 4.3 7 9.3 12 3.0
Psychologist contact
0 216 94.7 90 96.8 70 93.3 376 94.9
<1 8 3.5 1 1.1 3 4.0 12 3.0
1.0-2.9 3 1.3 0 0.0 1 1.3 4 1.0
3.0+ 1 0.4 2 2.2 1 1.3 4 1.0
Primary health care inpatient
treatment (F10)
0.003a
<0.01 p-r
0 104 45.6 43 46.2 53 70.7 200 50.5 <0.01 d-r
<1 93 40.8 30 32.3 15 20.0 138 34.8
1.0-1.9 18 7.9 10 10.8 4 5.3 32 8.1
2.0+ 13 5.7 10 10.8 3 4.0 26 6.6
Mental health care services
MH doctor visits <0.001b <0.001 d-r
0 183 80.3 85 78.9 49 65.3 317 80.1 <0.01 p-r
<1 39 17.1 7 7.5 13 17.3 59 14.9
1.0-1.9 5 2.2 0 0.0 7 9.3 12 3.0
2.0+ 1 0.4 1 1.1 6 8.0 8 2.0
MH nurse visits <0.001a <0.001 d-r
0 138 60.5 71 76.3 35 46.7 244 61.6 <0.05 p-r
<1 45 19.7 6 6.5 12 16.0 63 15.9
1.0-2.9 26 11.4 8 8.6 11 14.7 45 11.4
3.0-9.9 12 5.3 5 5.4 7 9.3 24 6.1
10+ 7 3.1 3 3.2 10 13.3 20 5.1
Specialized AUD services
AUD doctor visits <0.001b <0.05 d-r
0 143 62.7 80 86.0 52 69.3 275 69.4 <0.001 p-d
<1 71 31.1 10 10.8 12 16.0 93 23.5
1.0-1.9 10 4.4 1 1.1 8 10.7 19 4.8
2.0+ 4 1.8 2 2.2 3 4.0 9 2.3
ACCEPTED MANUSCRIPT
AUD nurse/socialwelfare
supervisor visits <0.001a <0.001 p-d
0 76 33.3 56 60.2 43 57.3 175 44.2 <0.05 p-r
<1 42 18.4 10 10.8 3 4.0 55 13.9
1.0-2.9 48 21.1 13 14.0 13 17.3 74 18.7
3.0-9.9 46 20.2 8 8.6 7 9.3 61 15.4
10+ 16 7.0 6 6.5 9 12.0 31 7.8
Social worker visits 0.084b <0.01 p-d
0 194 85.1 90 96.8 66 88.0 350 88.4
<1 18 7.9 2 2.2 4 5.3 24 6.1
1.0-1.9 11 4.8 1 1.1 2 2.7 14 3.5
2.0+ 5 2.2 0 0.0 3 4.0 8 2.0
AUD detoxification
treatment 0.013b <0.01 p-r
0 160 70.2 78 83.9 66 88.0 304 76.8 <0.05 p-d
<1 52 22.8 11 11.8 6 8.0 69 17.4
1.0-1.9 10 4.4 2 2.2 3 4.0 15 3.8
2.0+ 6 2.6 2 2.2 0 0.0 8 2.0
AUD rehabilitation
period 0.025b <0.05 p-d
Yes 38 16.7 6 6.5 4 9.3 48 12.1
No 189 82.9 87 93.5 68 90.0 344 86.9
AUD evaluation
period 0.214b
No 226 99.1 90 96.8 74 98.7 390 98.5
Yes 2 0.9 3 3.2 1 1.3 6 1.5
AUD housing
rehabilitation 0.766b
No 225 98.7 93 100.0 75 100.0 393 99.2
Yes 3 1.3 0 0.0 0 0.0 3 0.8
AUD interval care
period 0.699b
No 218 95.6 91 97.8 73 97.3 382 96.5
Yes 10 4.4 2 22.2 2 2.7 14 3.5
AUD crisis treatment
period 0.878b
No 223 97.8 92 98.9 74 98.7 389 98.2
Yes 5 2.2 1 1.1 1 1.3 7 1.8
AUD sobriety support
period 0.232b
No 223 97.8 93 100.0 75 100.0 391 98.7
Yes 5 2.2 0 0.0 0 0.0 5 1.3
Specialized health care
SPE doctor visits (F10) <0.001a <0.05 d-r
0 88 38.6 48 51.6 54 72.0 190 48.0 <0.001 p-r
<1 97 42.5 31 33.3 15 20.0 143 36.1
1.0-2.9 34 14.9 8 8.6 6 8.0 48 12.1
3.0+ 9 3.9 6 6.5 0 0.0 15 3.8
