Quality in Primary Care Open Access

Keywords

addiction, attitudes, drug dependence, primary care, quality

How this fits in with quality in primary care

What do we know?

The quality of care received by patients with alcohol and other drug-related problems can depend, in part, on clinicians’ attitudes, which may vary according to the patient’s substance dependence type (alcohol, other drugs, or both).

What does this paper add?

This paper shows that substance dependence type, particularly drug dependence (with or without alcohol dependence), and injection drug use, are associated with worse primary care quality in domains involving the physician–patient relationship.

Introduction

The Institute of Medicine (IOM)[1] and others[2,3] define primary care as sustained partnerships with essential attributes of comprehensiveness, coordination and accessibility. Many have advocated addressing alcohol and other drug (AOD) problems in primary care, as they are common issues in this setting.[4–6] Addictions are often chronic disorders,[7] and primary care has the potential to improve addiction outcomes. The quality of primary care delivered, including quality of care as reported by the patient, is linked to important patient care outcomes such as adherence to physician advice and improved health status.[8] In chronic disorders such as diabetes mellitus,[9] depression[10] and tobacco dependence,[8] primary care quality (PCQ) is an im-portant determinant of outcomes. Institutional, struc-tural, interactional and personal factors are associated with quality of primary care. Core features of PCQ, particularly those reflecting the quality of the phys-ician–patient relationship, are associated with positive addiction outcomes.[5]

Addressing AOD problems in primary care makes them mainstream health issues, thereby potentially decreasing stigma related to use of these substances and improving care. AOD-related stigma[11,12] may have direct consequences for the quality of care received by patients with AOD-related problems.[13,14] According to Fortney and colleagues,[15] the greater the feeling of stigmatisation perceived by alcohol users, the lower was the adherence to alcohol treatment. Injection drug users (IDUs) with HIV infection had significantly lower rates of exposure to antiretroviral therapy (13.5%) compared with patients with HIV who did not use injection drugs, who were cared for by the same physicians (36.1%).[16]

The quality of care received by patients with AOD-related problems can depend, in part, on clinicians’ attitudes.[17–19] Physicians may fear being deceived and manipulated by patients, not having the right tools to address AOD problems, and avoid talking about drugs with patients, even when this is the patient’s main concern.[20] Ronzani and colleagues[21] reported that health professionals seem to have less personal diff-culty working with patients who have alcohol depen-dence than they do with patients who have other substance dependence. Given these findings, and also given that some drug use is socially acceptable and even normative (e.g. alcohol use) and other use is highly stigmatised and generally unacceptable (e.g. injection drug use), it is likely that clinicians’ attitudes differ by type of substance use and disorder, and that in turn these attitudes could affect the quality of care they receive. However, the issues of clinicians’ atti-tudes differing by the type of substance use and these attitudes affecting quality of care, have not been largely studied.

Therefore, the aim of this study was to evaluate whether a patient’s substance dependence type (al-cohol, other drug, or both) was associated with PCQ. We studied patient self-reported PCQ and hypothe-sised that patients diagnosed with (other) drug de-pendence would have worse PCQ compared with alcohol dependence alone.

Methods

Study design

Data were obtained from a prospective cohort study of 563 alcohol- and/or drug-dependent patients enrolled in a randomised controlled trial – the Addiction Health Evaluation and Disease management (AHEAD) study. The AHEAD study was a trial of the effectiveness of alcohol and/or drug dependence chronic disease man-agement in primary care with 3-, 6- and 12-month in-person follow-up. The subjects were randomised to attend a chronic care management clinic (the AHEAD clinic (intervention) ) or to a control group that did not have access to the clinic. Both groups received appointments with a primary care physician and access to motivational enhancement counselling. The AHEAD clinic provided assessment, and ongoing care man-  agement on-site and by referral, by staff with experi-ence caring for patients with addictions (an internist, psychiatrist, social worker and nurse care manager). People eligible for the AHEAD study were adults (aged 18 years or older) with current alcohol dependence and heavy alcohol use and/or current drug depen-dence and recent opioid or stimulant use, who were able to interview and consent, speak English or Spanish, were willing to establish or continue primary care at the institution where the study took place, were willing to attend at least an initial AHEAD clinic visit and return for research interviews in follow-up. They had to provide two contact persons to assist with follow-up, have a Mini-Mental State Examination score > 20, and not be pregnant. Readiness to change substance use or seek help was not a criterion. Subjects were recruited from an in-patient detoxification unit, advertisements in the community, and clinical set-tings (primary care, hospital, emergency department) at an urban safety-net (i.e. underserved community, cared for all regardless of ability to pay) medical centre.

 To study the association between substance depen-dence type and primary care quality, we studied subjects from the AHEAD trial who had exposure to primary care, which was assessed in follow-up inter-views. As such, subjects in this analysis completed at least one follow-up interview. In addition, only inter-views in which subjects reported having a primary care clinician (PCC; affrmative response to ‘Is there one particular doctor (or primary care provider, e.g. Nurse Practitioner or Physician Assistant) that you consider to be your regular personal primary care doctor?’) were included in the current analyses. Subjects pro-vided written informed consent and were assessed in person by trained research assistants using stand-ardised interviews, and with assurances of confiden-tiality. Interviews covered demographics, substance use and severity, readiness to change drinking and drug use, mental health, and healthcare utilisation. We also assessed whether they had a PCC at each timeframe, and for how long – primary care utilisation (number of visits to the doctor in the past 3 months) and primary care duration (< 6 or _ 6 months). They received modest compensation for their time. The study was approved by the Institutional Review Board at the Boston University Medical Campus. A certifi-cate of confidentiality issued by the federal govern-ment further protected subjects’ privacy.

Main independent variable

The main independent variable of interest was sub-stance dependence type as assessed and determined by the Composite International Diagnostic Interview-Short Form[22–24] at study enrolment. Subjects were classified into one of the following three categories: current (past-year) alcohol dependence, drug depen-dence, or both alcohol and drug dependence.

As a secondary analysis, because we suspected that the greater stigma associated with injection drug use (versus drug dependence in general) might have a greater effect on PCQ, we replaced substance depen-dence type with any injection drug use (the question asked was: ‘In your lifetime, have you ever injected drugs?’).[25]

Dependent variables

The Primary Care Assessment Survey (PCAS)[8] was administered during follow-up interviews. It is a well-validated, patient-completed survey that measures defined attributes of primary care described by the Institute of Medicine (1996).[1–3,26] The PCAS is a measure of PCQ in the context of a specific physician– patient relationship. Nine summary scales (Table 1) cover two broad areas of PCQ: the patient–physician relationship (communication, interpersonal treatment, thoroughness of the physical exam, whole-person knowledge, preventive counselling, and trust) and structural/organisational features of care (organisa-tional access, financial access, and visit-based conti-nuity). Each scale has a range from 0 to 100 with higher scores indicating more of the underlying attribute. Because we anticipated most subjects would have primary care at one institution where the structural/ organisational features of care would vary very little, and because we were primarily interested in how dependence type would affect the physician–patient relationship, we analysed only the six patient–phys-ician relationship summary scales.

Table 1: Summary of item content for Primary Care Assessment Survey – range from 0–100, where 100 is best (adapted from Safran et al, 1998 and Kim et al, 2007)

Covariates

Potential confounders were selected based on the literature[5,10,27] and clinical knowledge. They included age, sex, race/ethnicity (white vs. other), the mental component summary (MCS) scale of the Short Form Health Survey (SF-12),[28] homelessness (one or more nights in a shelter or on the street in the past 3 months), employment (any vs. none), education (high school or more vs. less), any medical comorbidity, income ($19 999 or less, $20 000–$49 999, and $50 000 or more), health insurance (any vs. none), incarceration (ever vs. never), and randomisation group (of note, in analyses of the randomised trial results, the inter-vention was not associated significantly with sub-stance use or other health outcomes).[29]

Analysis

We generated descriptive statistics of all covariates at baseline stratified by dependence type. In addition, we described primary care utilisation across the follow up period (i.e. at 3, 6 and 12 months post baseline) using two different variables. We estimated the me-dian number of primary care visits in the past 3 months reported across follow-up visits, and also the proportions reporting they had a PCC for more than 6 months at each visit.

The association between substance dependence type (or injection drug use) and PCQ outcomes was tested using the general linear model for correlated data.[30] The models incorporated outcomes from the same subject at multiple timepoints (3, 6 and 12 months) and use all available observations from each subject (i.e. can handle incomplete data due to missing responses). The models used an unstructured covari-ance matrix (i.e. no explicit structure was imposed on the covariance among repeated measurements), where each variance and covariance between pairs of re-peated measures was estimated using all available data. Analyses adjusted for baseline values of age, sex, race/ethnicity, randomisation group, income, comor-bidity and education, and the following were treated as time-varying covariates (updated at each time point): mental component summary, homelessness, employ-ment, health insurance and incarceration. To minimise the potential for collinearity, we assessed correlation between pairs of independent variables and covariates to verify that no pair of variables included in the same regression model was highly correlated (i.e. r > 0.40). Analyses were conducted using two-sided tests and a significance level of 0.05. Owing to the exploratory nature of the analyses, adjustments were not made for multiple comparisons. To aid in the interpretation of results, adjusted mean PCQ scale values by dependence type were generated for the 3-month timepoint, sub-stituting overall sample means for continuous covariates and observed proportions for categorical covariates. All analyses were conducted using SAS Software (version 9.2; SAS Institute, Cary, NC).

Results

Of 427 subjects included in the study sample (i.e. those in the AHEAD study, who had at least one follow-up interview and reported having exposure to primary care at any time during follow-up), 299 reported having a PCC at the first follow-up (3-month) inter-view, 314 reported having a PCC at 6 months and 345 reported having a PCC at 12 months, contributing to a total of 958 observations in the longitudinal regression models. Sixty-one per cent, 71% and 83% of the subjects reported having a PCC for more than 6 months at the 3-, 6- and 12-month follow-ups, respectively. The median (IQR) total number of primary care visits across follow-up was 1 (0, 2) for those with one follow-up visit (data represent total visits in 3-month time period); 2 (1, 4) for those with two follow-up visits (data represent total visits in 6-month time period); 3 (2, 6) for those with three follow-up visits (data represent total visits in 9-month period).

In our sample, 14% of the respondents had alcohol dependence only, 20% had drug dependence only, and 66% had both alcohol and drug dependence. The sociodemographic and health characteristics of the study sample are displayed in Table 2. Subjects with alcohol dependence were older (mean age 46 years) than those with both alcohol and drug dependence (mean age 40 years) and those with drug dependence alone (mean age 34 years). Most of the participants were men, unemployed, and had been incarcerated before. A majority of study subjects (more than 80%) were insured and a substantial minority (less than 25%) had not graduated from high school. Those with alcohol dependence only were more likely to have medical comorbidity and less likely to have used injection drugs. However, those with drug depen-dence only had the highest prevalence of ever injecting drugs and were least likely to be homeless and have health insurance. Those with both alcohol and other drug dependence were the most likely to ever have been incarcerated.

Table 2: Study sample characteristics stratified by substance dependence type* at baseline (n = 427)

In adjusted analyses, dependence type was signifi-cantly associated (global P < 0.05) with patient self-reported PCQ for all patient–physician relationship PCAS scales except whole-person knowledge (Table 3). In pair-wise comparisons, subjects with alcohol dependence rated the quality of primary care that they received significantly higher than those with drug dependence only or both alcohol and drug depen-dence in four domains: thoroughness of physical exams, communication, interpersonal treatment and trust. Those with drug dependence only rated their preventive care significantly worse than those with both alcohol and drug dependence. Mean adjusted differences ranged in absolute value from 0.02 to 10.21 points on these 100-point summary scales. Adjusted mean scores for each of the physician–patient re-lationship PCAS scales for the sample at 3 months for alcohol dependence only, drug dependence only, and both (alcohol and drug dependence), respectively, were: thoroughness of exams (76, 68 and 70), com-munication (80, 73 and 74), interpersonal treatment (79, 71 and 71), whole-person knowledge (62, 55 and 58), trust (80, 72 and 72) and preventive counselling (60, 57 and 67).

Table 3: Association between substance dependence type and primary care quality across time (as measured by PCAS scale scores (adjusted mean differences) )

Concerning the associations between injection drug use and primary care quality, in adjusted analyses, injection drug use was significantly associated with lower primary care quality in five of the six physician– patient-relationship domains (all but preventive counselling; see Table 4).

Table 4: Association between injection drug use and primary care quality across time, as measured by the PCAS scale scores (adjusted mean differences)

Discussion

In this study, having drug dependence, with or with-out alcohol dependence, was associated with receiving significantly worse patient-rated quality of primary care across most (four of six) domains involving the physician–patient relationship. Injection drug use specifically, a highly stigmatised behaviour, was con-sistently associated with lower quality of primary care. These associations were present in analyses adjusted for potential confounders such as sex, race/ethnicity and health insurance.

People with AOD-related problems can get poor quality of care in part as a result of clinicians’ attitudes, including stigmatising attitudes,[11,12,31] which may help to explain the disparities in quality of care found in this study. Stigmatisation of people with drug-related problems is a historical, political and social phenomenon[32] from which health professionals are not immune. In Australia, IDUs reported being dis-criminated against by the police (80%), hospital staff (60%), doctors and pharmacists (57%), methadone providers (33%) and community health workers (7%).[33] In addition to stigmatisation as an explanation for differences in quality of care, other studies have demonstrated the importance of the physician– patient relationship (a part of quality primary care), for the receipt of high-quality care and for better outcomes.[10,34] Among HIV-infected individuals, trust

This study had some limitations that should be considered in interpreting the results. Causality can-not be determined among variables due to the study design – a prospective cohort study. Furthermore, it is possible that the duration of the primary care relation-ship (which we did not test because it could be in the causal pathway between substance dependence type and quality of care) is important to quality, and results could be different if we had studied longer-term relationships or those with more frequent visits. The generalisability of the study’s findings may be limited to an urban adult population with social and econ-omic problems, an important group of people with substance dependence. Finally, we studied patient self-reported PCQ. There are a number of ways to measure PCQ, from self-report to observation of processes of care, and even outcomes of care. We chose one method, the PCAS, because it is comprehensive, reflects the definition of primary care promulgated by the Institute of Medicine, and because for substance dependence, the physician–patient relationship is likely what is most at risk. Furthermore, scores on the PCAS have been associated with clinical out-comes[5,8] and with changes in primary care performance in longitudinal studies (physician’s knowledge of patient and trust).[3]

Efforts to improve healthcare for people with sub-stance dependence need to go beyond simply provid-ing good medical care or training health professionals who are going to deliver it.[17] New strategies should hold the potential to make addictions mainstream health and social conditions. We speculate that such a change will need to be based on multipronged efforts across a range of areas in society. Eventually such a focus may lead to decreased stigma and improved services for patients with substance dependence who receive primary medical care.

In addition, we suggest additional avenues for research. Future studies should examine the relation-ship between quality of care and important clinical outcomes in patients with substance dependence. Such studies would be preludes to the development of interventions that improve quality of care and subsequent outcomes in these patients.

In conclusion, it appears that substance dependence type, particularly drug dependence (with or without alcohol dependence), and injection drug use, are associated with worse-quality primary care for several domains, among people with substance dependence. The clinical significance of these findings should be viewed in light of the fact that more understanding is needed about the reasons for the differences found here. Addressing them may be important for improv-ing the quality of primary care for patients with addictions.

Acknowledgements

We presented preliminary results of this work as a poster during the 2009 Addiction Health Services Research Conference in San Francisco, California.

References

1. Institute of Medicine. Primary Care: America’s health in a new era. Washington, DC: National Academy Press, 1996.
2. Starfield B, Cassady C, Nanda J, Forrest CB and Berk R. Consumer experiences and provider perceptions of the quality of primary care: Implications for managed care. J Fam Practice 1998;46:216–26.
3. Safran DG. Defining the future of primary care: What can we learn from patients? Ann Intern Med 2003; 138:248–55.3.
4. Humphreys K and Tucker JA. Toward more responsive and effective intervention systems for alcohol-related problems. Addiction 2002;97:126–32.
5. Kim TW, Samet JH, Cheng DM, Winter MR, Safran DG and Saitz R. Primary care quality and addiction severity: A prospective cohort study. Health Services Research 2007;42:755–72.3.
6. Saitz R, Larson MJ, LaBelle C, Richardson J and Samet JH. The case for chronic disease management for addiction. J Addict Med 2008;2:55–65.
7. McLellan AT, Lewis DC, O’Brien CP and Kleber HD. Drug dependence – a chronic medical illness: implications for treatment, insurance, and outcomes evaluation. JAMA 2000;284:1689–95..
8. Safran DG, Kosinski M, Tarlov MA et al. The Primary Care Assessment Survey: Tests of data quality and measurement performance. Med Care 1998;36:728–391.
9. Campbell S, Roland M, and Wilkin D. Primary Care Groups: Improving the quality of care through clinical governance. BMJ 2001;322:1580–2.
10. O’Malley AS, Forrest CB and Miranda J. Primary care attributes and care for depression among low-income African American women. Am J Public Health 2003;93: 1328–34.
11. Corrigan PW, Lurie BD, Goldman HH, Slopen N, Medasani K and Phelan S. How adolescents perceive the stigma of mental illness and alcohol abuse. Psychiatr Serv 2005;56(5):544–50
12. Room R. Stigma, social inequality and alcohol and drug use. Drug Alcohol Rev 2005;24:143–55.
13. Olsen JA, Richardson J, Dolan P and Menzel P. The moral relevance of personal characteristics in setting health care priorities. Soc Sci Med 2003;57:1163–72.
14. Stone VE, Weissman JS and Cleary PD. Satisfaction with ambulatory care of persons with AIDS: predictors of patient ratings of quality. J Gen Intern Med 1995; 10(5):239–45.1
15. Fortney J, Mukherjee S, Curran G, Fortney S, Han X and Booth B. Factors associated with perceived stigma for alcohol use and treatment among at-risk drinkers. J Behav Health Ser R 2004;31:418–29
16. Ding L, Landen B, Wilson I, Wong M, Shapiro M and Cleary P. Predictors and consequences of negative physician attitudes toward HIV infected injection drug users. Arch Intern Med 2005;165:618–23
17. Druss BG, Marcus SC, Rosenheck RA, Olfson M, Tanielian T and Pincus HA. Understanding disability in mental and general medical conditions. Am J Psychiat 2000;157(9):1485–91
18. Palm J. The nature of and responsibility for alcohol and drugs problems: views among treatment staff. Addict Res Theory 2004;12(5):413–31
19. McCormick KA, Cochran NE, Back AL, Merrill JO, Williams EC and Bradley KA. How primary care providers talk to patients about alcohol -Aqualitative study. J Gen Intern Med 2006;21:966–72
20. Merril JO, Rhodes LA, Deyo RA, MarlattGAand Bradley KA. Mutual mistrust in the medical care of drugs. The keys to the ‘‘Narc’’ cabinet. J Gen Intern Med 2002; 17:327–33
21. Ronzani TM, Higgins-Biddle J and Furtado EF. Stigmatization of alcohol and other drug users by primary care providers in Southeast Brazil. Soc Sci Med 2009;69:1080–4
22. Kessler RC, Andrews G, Mroczek D, Ustun TB, Wittchen and H-U. The World Health Organization Composite International Diagnostic Interview Short Form (CIDISF). Int J Method Psych 1998;7(4):171–85
23. Kessler RC and Ustun TB. The World Mental Health (WMH) Survey initiative version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI). Int J Method Psych 2004;full text
24. Kessler RC, Abelson J, Demler O et al. Clinical Calibration of DSM-IV Diagnoses in the World Mental Health (WMH) Version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI), 2004
25. Metzger DS, Navaline HA and Woody GE. Assessment of Substance Abuse: HIV Risk Assessment Battery (RAB). Farmington Hills, MI: Macmillan, 2001.
26. Kroenke K. The many C’s of primary care. J Gen Intern Med 2004;19:708–9
27. Saitz R, Horton NJ, Larson MJ, WinterMand Samet JH. Primary medical care and reductions in addiction severity: A prospective cohort study. Addiction 2005; 100(1):70–8.
28. Ware JE. SF-36 Physical and Mental Health Summary Scales: a user’s manual. Boston, MA: The Health Institute, 1994
29. Saitz R, Witas J, Meli S et al. Effectiveness of chronic care management for alcohol and drug dependence: Addiction Health Evaluation and Disease Management (AHEAD) trial results. Alcoholism: Clin & Exp Res 2010;34(s2):65A.
30. Fitzmaurice GM, Laird NM and Ware JH. Applied Longitudinal Analysis. Hoboken: Wiley, 2004
31. Amaral-Sabadini MB, Saitz R and Souza-Formigoni MLO. Do attitudes about unhealthy alcohol and other drug (AOD) use impact primary care professionals0 readiness to implementAOD-related preventive care? Drug Alcohol Rev 2010;29:655–61
32. Grimberg M. ‘‘Estigmatizacio´n y discriminacio´n social relacionada al VIH en pai´ses de Ame´rica Latina: sexualidad y uso de drogas en jo´ venes en contextos de precarizacio´n social urbana’’. In: Estigma y discriminacio ´n por el VIH sida: un enfoque antropolo´gico, Estudios e Informes, Serie Especial (20). Divisio´n de Poli´ticas Culturales y Dia´logo Intercultural, UNESCO, 2003.
33. Burrows D. Stigmatisation of HIV-positive injecting drug users. In: Stigma-AIDS Online Forum, Stigma, HIV and Drug Users-3, https://archives.healthdev.net/ stigma-aids/msg00004.html. Cited in: Foreman M, Lyra P, Breinbauer C. Comprensio´n y respuesta al estigma y a la discriminacio´n por el VIH/sida en el sector salud. Organizacio´n Panamericana de la Salud, 2003.
34. O’Malley AS, Sheppard VB, Schwartz M and Mandelblatt J. The role of trust in use of preventive services among low-income African-American women. Prev Med 2004;38:777–85.
35. Schneider J, Kaplan SH, Greenfield S, Li W and Wilson IB. Better physician–patient relationships are associated with higher reported adherence to antiretroviral therapy in patients with HIV infection. J Gen Intern Med 2004;19:1096–103
36. Preau M, Leport C, Salmon-Ceron D, et al. Healthrelated quality of life and patient-provider relationships in HIV-infected patients during the first three years after starting PIcontaining antiretroviral treatment. AIDS Care 2004;16:649–61.

Funding

The AHEAD Study was supported in part by National Institute on Alcohol Abuse and Alcoholism (NIAAA) Grant Number R01AA010870 (Saitz, PI), National Institute on Drug Abuse (NIDA) Grant Number R01DA010019 (Samet, PI) and National Center for Research Resources (NCRR) Grant Number UL1RR025771 (Center, PI). Michaela Bitarello do Amaral-Sabadini had financial support from Coordenac¸a˜o de Aperfeic¸oamento de Pessoal de Nı´vel Superior (CAPES) (Scholarship – Process number: # 0389099).

Ethical Approval

The study was approved by the Institutional Review Board at the Boston University Medical Campus. A certificate of confidentiality issued by the federal government further protected subjects’ privacy. The material has not been published in whole or in part elsewhere. The paper is not currently being considered for publication elsewhere.

Peer Review

Not commissioned; externally peer-reviewed.

Conflicts of Interest

To the best of our knowledge, no conflict of interest, financial or other, exists. The content of this manu-script is solely the responsibility of the authors and does not necessarily represent the offcial views of the National Institute on Drug Abuse, the National Insti-tute on Alcohol Abuse and Alcoholism, the National Center for Research Resources, the National Institute of Health or the Coordenac¸a˜o de Aperfeic¸oamento de Pessoal de Nı´vel Superior. These institutions had no further role in study design, in the collection, analysis and interpretation of data, in the writing of the report, or in the decision to submit the paper for publication.