Chronic illness and multimorbidity among problem drug users: a comparative cross sectional pilot study in primary care
© Cullen et al. 2009
Received: 21 October 2008
Accepted: 21 April 2009
Published: 21 April 2009
Skip to main content
© Cullen et al. 2009
Received: 21 October 2008
Accepted: 21 April 2009
Published: 21 April 2009
Although multimorbidity has important implications for patient care in general practice, limited research has examined chronic illness and health service utilisation among problem drug users. This study aimed to determine chronic illness prevalence and health service utilisation among problem drug users attending primary care for methadone treatment, to compare these rates with matched 'controls' and to develop and pilot test a valid study instrument.
A cross-sectional study of patients attending three large urban general practices in Dublin, Ireland for methadone treatment was conducted, and this sample was compared with a control group matched by practice, age, gender and General Medical Services (GMS) status.
Data were collected on 114 patients. Fifty-seven patients were on methadone treatment, of whom 52(91%) had at least one chronic illness (other then substance use) and 39(68%) were prescribed at least one regular medication. Frequent utilisation of primary care services and secondary care services in the previous six months was observed among patients on methadone treatment and controls, although the former had significantly higher chronic illness prevalence and primary care contact rates. The study instrument facilitated data collection that was feasible and with minimal inter-observer variation.
Multimorbidity is common among problem drug users attending general practice for methadone treatment. Primary care may therefore have an important role in primary and secondary prevention of chronic illnesses among this population. This study offers a feasible study instrument for further work on this issue. (238 words)
Injecting drug users have an increased mortality compared to the general population. Mortality rates of 13–30 per thousand persons per year (an age-adjusted mortality rate of 6.9–22.0) have been reported among injecting drug users, with opiate overdose and HIV infection the major causes of death [1–5].
The introduction of 'harm reduction' interventions (ie interventions designed to prevent problems associated with problem drug use, eg oral methadone treatment, needle and injecting paraphernalia exchange programmes) has, in addition to advances in HIV treatment, been an important factor in reducing mortality/increasing life expectancy among problem drug users [6–8]. One might therefore expect an 'epidemiological shift', as chronic illnesses and their complications replace infectious diseases and drug overdose as causes of death. In this respect, an association between problem drug use and subsequent decrease in general self-rated health has been demonstrated .
While bloodborne virus infections (in particular hepatitis C [10–12]), psychiatric illness  and to a lesser extent, problem alcohol use  are all recognised as chronic medical problems that are more prevalent among injecting drug users, our understanding of other 'common' chronic illnesses among injecting drug users is less advanced. While much is known about the epidemiology and management of diabetes, cardiovascular disease and chronic respiratory illness (eg asthma, COPD) and while the prevention and management of these chronic illnesses has been identified as a priority issue for population health in Ireland and globally [15, 16], no data have reported on the epidemiology or care of these illnesses among current or former drug users.
- to determine chronic illness/multimorbidity prevalence and health service utilisation among problem drug users attending primary care for methadone treatment and to compare these rates with matched 'controls';
- to develop and pilot test a valid study instrument.
The study was carried out in Dublin city, where an estimated 16.0 per thousand of the population currently use illicit opiates . In Ireland, addiction treatment is currently provided by specialist addiction treatment services, including a central Drug Treatment Centre Board, regional addiction centres, community-based projects (satellite clinics) and by primary care. Most recent data reported that 7845 patients were treated for problem drug use by these agencies in 2002, with opiates the most common drug for which people attended for treatment (86% of total) .
In recent years, the number of general practitioners (GPs) prescribing methadone has increased in Ireland, the UK and elsewhere in the EU [19–23]. To prescribe methadone in Ireland, GPs must complete special training and are subject to clinical audit, with GPs who provide methadone treatment for 15 or more patients subject to more regular audit and advanced training ('level 2' GPs) . This system is analogous to the 'GPs with a special interest' ('GPWsi') model currently operating in the UK . Initiation of methadone treatment is only permitted in specialist addiction clinics or by 'level 2' GPs . In all other cases, chaotic opiate users are initially cared for in a specialist addiction clinic and their care then transferred to general practice when stable. In circumstances where illicit drug use of patients attending general practice becomes 'chaotic', addiction care is transferred to the specialist addiction services.
Characteristics of practices that participated in study.
Approximate number of patients on methadone treatment at practice
3.0 FTE doctor teaching general practice in Dublin's south inner city
Four doctor teaching general practice in Dublin's south inner city
2.5 FTE general practice in Dublin's north inner city
All patients attending each practice for methadone treatment on 1st July 2008 were eligible for the study. A considerable variability in numbers of patients attending the three practices for methadone existed at the time of the study, with 19 the smallest number of patients attending one of the practices for methadone treatment and hence 19 cases were sampled at each practice.
At each participating practice, the researcher and GP identified cases for inclusion in the study by random number sampling from a numbered list of patients attending that practice for methadone treatment (in Ireland, GPs are not allowed to issue a prescription for methadone for any more than two weeks and are remunerated on a 'per consultation' basis).
Controls were matched for practice, age, GMS scheme membership and gender. Potential controls were identified by running a query of 'active patients' registered on each practice's electronic practice management information system using the matching criteria. This list of potential controls was then numbered and one control identified using random number sampling.
Data was collected from clinical records. A study instrument was developed for the purpose of this study (see Appendix 1). The instrument was based on another instrument routinely used to collect health information on current or former injecting drug users in Ireland, the Health Research Board National Drug Treatment Reporting System (NDTRS) . Data were therefore collected on:
Presentations with acute illnesses
Health service utilisation (including: attendance at primary care, secondary care, out of hours services, emergency departments and diagnostic investigations).
Data were collected by four researchers over the course of two weeks in July 2008: one researcher (SOB) collected data on all records and at each practice, one of the GPs at that practice (WC, FDOK or AOC) also collected data. Data was extracted from the 'diagnosis' 'prescribing', 'consultation', and 'communication' modules from individual electronic clinical records. In addition, paper records were reviewed for: 'active diagnoses/chronic illnesses', and hospital referral/update/discharge letters.
At consecutive intervals, the researchers compared the data collected to determine the degree of inter-observer variability; this check was conducted after six, 12, 19, 38 and 57 records of patients on methadone had been reviewed. At each review, questions with an inter-observer agreement of less than 90% and questions where data extraction had proved problematic were revised and these revisions were incorporated into subsequent iterations of the study instrument. A log of the data collection process was maintained, with the amount of data collected in one three-hour sessional block and problems in data collection recorded.
Data were analysed using Statistical Packages for the Social Sciences (SPSS) version 12.0. Analytical techniques included Pearson's chi squared test to determine the significance of associations between categorical variables. Odds ratios and their 95% confidence intervals (CI) were used to compare rates of chronic and acute illness and health service utilisation between the two groups (cases and controls).
All data were collected and/or reviewed by a GP with clinical responsibility for the patient's ongoing care. The strategy whereby GPs were involved in data collection was adopted for two reasons: to optimise data accuracy and in case any health issue was identified that may require a subsequent clinical intervention. Any subsequent clinical interventions were not included in the data collection.
- all data were collected anonymously and any details that could potentially identify individuals were removed after data collection;
- all members of the research team were nominated as 'agents' of each practice and each signed a confidentiality agreement with the other participating practices.
Description of data collection process from clinical records of patients being prescribed methadone.
Items with inter-observer disagreement/total
Workload (in number of three-hour sessions) involved in data collection to collect data
The mean age of 'cases' was 37.2 years (standard deviation: 7.25 years), 42(74%) were male, 41(72%) had GMS cover, all were Irish nationals and 56(98%) were documented as living in stable accommodation.
Reasons why, and time since, patients on methadone first attended practice.
Reason first attended practice
First attended the practice <10 years ago
First attended the practice >10 years ago
Treatment of illicit drug use
Referred by addiction services for methadone treatment
General medical care
Considerable contact with specialist addiction services (both lifetime and since commencing methadone treatment at the practice) was also observed among the sample. All patients had been prescribed methadone by specialist addiction services prior to attending the general practice for methadone treatment and 14(25%) patients had been referred back to specialist addiction services for methadone treatment since commencing methadone treatment at the practice (six in the previous year).
The mean daily dose of methadone prescribed (on the last issued prescription) was 66 milligrammes (standard deviation: 26 milligrammes). Tobacco/cigarette smoking status was recorded in the records of 39 patients (68% of total), of whom 37 were recorded as smokers.
Prevalence of, and attendance at secondary care for specific chronic illnesses same.
Chronic illness (ICPC code)
Number of patients with illness documented/number who have attended secondary care for this illness (cases)
Diabetes-Insulin Dependent (T90)
Diabetes – Non Insulin Dependent (T91)
Heart valve disease (K83)
Chronic obstructive pulmonary disease (R95)
Viral hepatitis (D72): hepatitis C
Viral hepatitis (D72): hepatitis B
Anxiety Disorder (P74)
Chronic alcohol abuse (P15)
Other chronic illnesses documented
Thirty-one patients (54%) had an acute illness during the previous three-months, (mean acute illnesses per patient: 1.0, standard deviation: 1.5). Upper respiratory tract infection (10 patients), sleep disturbance/anxiety/feeling depressed (four), abdominal pain (four), urinary tract infection (three) and ear wax (three) were the most commonly documented acute illnesses.
Twenty-five patients (44% of total) had been prescribed at least one time-limited medication (in addition to methadone) for the treatment of an acute illness during the previous three months (mean medications per patient: 0.9, standard deviation: 1.5).
In the previous six months, patients on methadone had attended their GP for issues other than their addiction care a mean of 3.9 times (standard deviation: 4.1), and had attended another healthcare professional in the practice a mean of 0.5 times (standard deviation: 0.8). Three patients had used the out of hours services provided by the practice and ten patients (18%) had at least one diagnostic investigation arranged or performed by the practice. A total of 23 such investigations were performed on these ten patients, which included: biochemistry/haematology (13), microbiology (six), cervical smears (three) and radiology (one).
In the previous six months, 27 patients (47%) had either been referred to, or attended secondary care, with Emergency Department (10 referrals/attendances), infectious diseases (10 referrals/attendances), gastroenterology (8 referrals/attendances), hepatology (6 referrals/attendances), addiction services (4 referrals/attendances) and psychiatry (3 referrals/attendances) the secondary care services to which patients were most commonly referred/attended.
The mean age of 'controls' was 37.2 years (standard deviation: 0.44), 42(74%) were male, 41(72%) had GMS cover and all were Irish nationals. Thirty-eight (67%) were documented as living in stable accommodation.
Among the 'control' group (n = 57), morbidity and health service utilisation rates were also high, with 40(70%) having a documented chronic illness and 23(40%) being prescribed recurrent medications. In the previous six months, 27(47%) had consulted with a healthcare professional at the practice, 19(33%) were referred to secondary care and 11(19%) had investigations performed by the practice. In the previous three months, 21(37%) had attended with an acute illness.
Patients attending for methadone treatment compared to randomly sampled population matched by practice, age, gender and GMS cover.
Patients on methadone
Randomly sampled control population
Odds ratio (95% confidence interval)
Chi squared (p value)
On recurrent medications
Attendance with acute illness a
Medication prescribed for acute illness a
Attended GP/healthcare professional at practiceb
Referred to/attended secondary care b
Attended practices' out of hours/deputising service b
Has had investigations performed/arranged by practice b
The proportions of cases (patients on methadone) and controls that had a chronic illness, that were prescribed regular medication, that attended with acute illnesses and that consulted with primary/secondary care were high, although patients on methadone treatment were more likely to have at least one chronic illness recorded, to be prescribed regular medications and to have attended primary care.
Validity of the data reported in this paper is likely to have been enhanced by the practices in which it was conducted (research-active, with advanced practice information systems) and the method of data collection (two researchers validating and cross-checking data that had been extracted and a study instrument developed to minimise variation between researchers).
However, we acknowledge a number of potential sources of bias. Participating practices had a longstanding clinical and research interest in problem drug use. While not representative of practices providing methadone treatment in Dublin, we considered them an appropriate environment for exploratory research on this subject.
Ascertainment bias is also possible as patients on methadone must attend their GP every 1–2 weeks in Ireland . As data were collected from clinical records, it is possible that frequent attendance (for methadone treatment) may increase documentation of chronic illnesses. It is also possible that this review of clinical records may have under/over reported other behaviours such as smoking. Selection bias was also possible, as patients with less severe and extreme problem drug use do not attend general practice for methadone treatment in Ireland.
Such considerations notwithstanding, the sample was comparable to larger samples of patients attending general practice for methadone treatment in Ireland in terms of gender and bloodborne virus status, but was older [27, 28].
Work conducted in a similar setting to ours has identified a number of barriers to multimorbidity research and these include: problems with practice software, variations in disease coding and accurately determining primary and secondary care activity through clinical records . While our experience would support these barriers and while morbidity data extracted from clinical records will inevitably be determined by clinical record keeping, we suggest that adopting a study instrument that is already being used in primary care  and modifying this to minimise inter-observer variation, can yield consistent data across practices.
Our findings regarding multimorbidity and chronic illness among patients on methadone treatment are consistent with North American studies that have reported higher prevalence of medical and psychiatric conditions among problem substance users [30, 31]. Our findings regarding health service utilisation is consistent with other work which show illicit drug users more likely to use Emergency Department and primary care .
While chronic respiratory disease may not yet be widely recognised as such, this study highlights the potential importance of its prevention and treatment among problem drug users. We anticipated the prevalence of other chronic illnesses (eg chronic alcohol abuse, diabetes, hypertension, ischemic heart disease) might have been higher and reasons for this and incidence rates should be explored in further research, especially as the cohort ages. In time, it is possible that adverse lifestyle factors will increase the incidence of these chronic illnesses, thereby leading to more complex care.
This study has also documented a high prevalence of chronic illness, incidence of acute illness and high contact rates with primary/secondary care among 'controls', findings which may be explained by the study being conducted in areas of high deprivation .
This pilot study has highlighted a need for further research on the epidemiology of chronic illnesses among patients on methadone. A larger, more representative sample of practices would make for more generalisable findings regarding illness prevalence. Conducting such research at practices from a range of socio-economic areas would allow controlling for deprivation as a potential confounder. Longitudinal studies would enable determination of incidence of key chronic illnesses, with data being collected directly from patients as well as from clinical records.
Such research should explore opportunities for primary and secondary prevention of chronic illness and determine uptake of primary care interventions in chronic illness management.
In this pilot study, a 21% difference in chronic illness prevalence rates between 'cases' and 'controls' was documented. However, chronic illness prevalence rates among problem drug users and 'controls' may have been higher in our study for reasons discussed above. Allowing for a more conservative estimate of chronic illness prevalence among (general) young adult populations of 48% , a 21% difference among problem drug users, we estimate data on 86 patients on methadone and 86 controls would determine a statistically significant difference in chronic illness prevalence rates (assuming significance of 0.05, power of 0.90 and a case:control ratio of 1:1).
If the findings of this pilot study are supported by more powerful/representative studies, then screening and treatment of chronic illness and increasing care complexity will be important issues in the future management of problem drug users. Therefore, an integrated care model, in which primary care and addiction care both care for problem drug users may best address this population's health needs .
Multimorbidity is common among problem drug users attending general practice for methadone treatment, and primary care may have an important role in primary and secondary prevention of chronic illnesses among this group. Further work on chronic illness and health service utilisation among problem drug users is advocated and this study offers a feasible study instrument.
HSE area of practice
Type of addiction treatment
Being prescribed methadone at present: Yes No
Primary agency managing addiction problem: GP Addiction clinic
Client number ____________(Note deleted from final electronic file)
Gender Male Female
Age last birthday
Health cover GMS Non-GMS
GMS number (Note deleted from final electronic file)
Area of residence
County registration plate ______________
Date first referred to/attended general practice _____________
Main reason for first referral/attendance (tick)
Illicit drug use
Licit drug use
Referred by specialist addiction services for methadone treatment
General medical care/GMS registration
Was the patient prescribed methadone by any other agency/clinic before attending the practice for methadone treatment:
What is the earliest recorded date on which the patient was prescribed methadone by any agency/clinic (01/month/year) _____________
What is the earliest recorded date on which the patient was prescribed methadone at this practice (01/month/year) _____________
Has the patient attended any other agency/clinic for addiction treatment since the above date (21b)?
Has the patient attended any other agency/clinic for addiction treatment in the last 12 months?
Date last treatment episode with methadone commenced _____________
Dose of methadone at last prescription _________MILLIGRAMMES PER DAY
Which illnesses that require ongoing follow up have been documented at the time of data being collected and/or at any time in the past and in the case of each has patient attended secondary care for this problem
- Diabetes-Insulin Dependent (T90)
- Diabetes – Non Insulin Dependent (T91)
- Cardiac arrhythmia (K80)
- Rheumatic fever (K71)
- Ischaemic heart disease with angina (K74)
- Acute myocardial infarction (K75)
- Ischaemic heart disease (K76)
- Heart failure (K77)
- Pulmonary heart disease (K82)
- Heart valve disease (K83)
- Heart disease other (K84)
- Hypertension complicated (K87)
- Asthma (R96)
- Chronic obstructive pulmonary disease (R95)
- Depression (P76)
- Viral hepatitis (D72): hepatitis C
- Viral hepatitis (D72): hepatitis B
- HIV/AIDS (B90)
- Anxiety Disorder (P74)
- Chronic alcohol abuse (P15)
- Other chronic illnesses
Documented acute illnesses in the preceding 3 months (and their ICPC-2 code)
If the patient attended today for a repeat prescription, what medication would be prescribed? Medication (generic name)/route
Acute/non-recurrent medications in the last 3 months: medication (generic name)/route
In the last 6 months, number of consultations (excluding those concerning ONLY methadone) with:
- a doctor in the practice: ____________
- another healthcare professional in the practice: ____________
Has patient been referred to or attended secondary care (including emergency departments) in the last 6 months?
If yes, please specify: specialty/date referred by practice/date attended specialty
Has patient attended out of hours/GP deputising service in the last 6 months?
If yes, please specify: Date/Problem/ICPC code
Has patient had a diagnostic investigation arranged/performed by the practice in the last 6 months (excluding urine toxicology)? Yes No
If yes, please specify: pathology/diagnostic imaging/other
This study was funded by the UCD School of Medicine through its support of SO'B who participated in the Summer Students Research Awards Programme. We thank the GPs and practice staff for their assistance without which this study would not have been possible.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.