Patient compliance with prescribed medication regimens is a widespread problem. When it comes to medication itself, compliance can be broken down into taking the medication at the recommended dose, the correct frequency, the appropriate time, the appropriate way, and the appropriate duration. Besides taking medications as prescribed, compliance includes attending appointments and following various doctors' recommendations.
In many cases, patients have full capacity to make decisions regarding their healthcare, including a choice of whether or not they want to follow through with their doctor's recommendations. These patients should not be labeled as “noncompliant.” In the mental health area, the fine line between “noncompliance” and simple “refusal” is sometimes difficult to identify due to the inherent nature of many psychiatric disorders and the historical undermining of the rights of psychiatric patients.1 However, the consequences of noncompliance include increased rates of rehospitalization,5,7–9 longer hospital stays,5 increased economic burden,5 and poorer health outcomes.
This article will review the literature and present results from a new study regarding patient noncompliance with prescribed medications and follow-up appointments.
Measuring compliance is challenging. One might use regular measurements of medication blood levels, medication charting, the use of premanufactured pill packs, and pill counts.2,3 In addition, some methods used include checking pill bottles, reviewing patient's self-reports, assessing laboratory results, and electronic viewing of patient appointment records.4 Aside from blood draws, many of these different methods can be manipulated by the patients. Checking blood levels is not practical for most current psychotropic medications.
In one study of patients with severe mental illness, the rate of non-compliance with medications during a 1-year period was 33%.5 Other studies have shown that the rate of noncompliance averaged 42% with antipsychotic medications and 35% with antidepressant medications.6
Reasons for Noncompliance
Patient demographic factors. The literature is inconsistent on the relationship between demographic factors and compliance. Some studies have established a relationship between sex and compliance with psychiatric referrals, with women being more compliant than men,10 but other studies have not.11–13 Studies examining compliance with medications established sex as a factor, with women being more compliant than men.14 Some studies showed that age was related to compliance with psychiatric referrals, with older patients being more compliant,10,12 but others did not.11,13 Patients with higher educational levels and who were married were more likely to follow through with psychiatric referrals.11–13
Patient's attitude. The health belief model examines perceptions in four areas: vulnerability to illness, severity of the condition, efficacy and desirability of the treatment, and barriers to treatment.15 The health belief model is believed to play a major role in the patient's decision to carry through with specific recommendations.2,15
Social factors. Positive attitude of the patient's spouse and a good support system are related to higher compliance.2 Psychiatric medications are usually viewed less favorably and thought to have more side effects than medications used to treat physical conditions.16 This can affect compliance negatively.
Disease factors. Some studies did not find any correlation between the diagnosis itself and compliance.2,11 Others found that patients diagnosed with a personality disorder or substance abuse were less compliant than other patients.13 In some cases, patients stop their medications because of change in provider or remission of their condition.10,11
Treatment factors. The complexity of the medication regimen, cost, and side effects are important factors in compliance.2,3,17
Mental health provider factors. The importance of the doctor-patient relationship cannot be overemphasized in evaluating compliance. Proper communication, empathy, good rapport, and a professional relationship are all factors that increase compliance.2,3,18,19 Empowering patients through education about their illnesses and medications increases compliance.2,3,19
The time before discharge is crucial for interventions aimed at improving compliance.20 Interventions include education for the patients and their families, pillboxes and appointment reminders, empowering the patient and involving the support system, and keeping updated medication logs.3,4,20–22
Medication packages that are easy to open with simplified directions and pills that are easy to swallow help increase compliance.3 Cueing the action of taking the medication to another daily activity, such as brushing teeth, also can help foster compliance behavior.3,23
Merely distributing fact sheets of information to patients about their illness is not enough to improve post-discharge compliance. While handouts given to patients before discharge have been shown to improve their knowledge about the medications, compliance was only improved when the handout was accompanied by verbal reinforcement.24 A shorter period between day of discharge and the first after-care appointment and use of a referral coordinator who assisted patients with their post discharge appointments also can help increase compliance.25–27
Our staff possessed no information about factors related to compliance among adult patients discharged from our inpatient psychiatric unit. Identification of factors at time of discharge influencing future compliance might allow for modification of our standard of discharge care in order to improve long-range compliance and ultimately patient outcomes.
Based on the literature, we designed an intervention using a combination of measures that had been reported as effective in improving compliance. Here we report the results of our study comparing our standard of discharge care to the intervention focusing on compliance with aftercare — defined as compliance with medications and follow-up appointments following hospital discharge.
Eligible subjects were drawn from our adult inpatient unit (18 and older) at the Cleveland Clinic Foundation in Cleveland, OH. All patients being discharged from the unit between July 8, 2003, and September 9, 2003, were approached to participate in the study. Patients being transferred to another psychiatric facility or to another section of the hospital and patients with mental retardation or dementia were excluded. Seventy-one inpatients about to be discharged met the inclusion criteria and were approached to participate in the study. Fifteen patients refused to participate in this study, and 56 enrolled.
Initial and follow-up questionnaires were used. Questions in the initial questionnaire were created from the variables that previous studies had found to be related to non-compliance. Table 1 (see page 166) summarizes the questions used in the initial questionnaire. Table 2 (see page 169) summarizes those in the follow-up questionnaire.
Summary of Items on Initial Questionnaire
Summary of Items on Follow-up Questionnaire
The follow-up questionnaire examined compliance from three different perspectives. First, the patient was asked to give a self-report on adherence to the prescribed medication regimen and follow-up appointments. The patient's prescribing physician and pharmacy were contacted independently to provide an objective measure of the patient's compliance. In the follow-up questionnaire, the patient was asked about current mental health status and reasons for noncompliance when applicable. Patiens were who receive the intervention were also asked to provide an evaluation.
After informed consent was obtained, patients in the study were randomized into two groups. Group A consisted of subjects who received only the hospital's treatment as usual (TAU) discharge planning. This included prescriptions for their medications and a discharge sheet that lists their diagnosis, medications, and follow-up appointments. These sheets were distributed and read to the patients by a nurse immediately preceding discharge.
Group B received the hospital TAU in addition to the study's intervention. The intervention was conducted by a physician during a 30-minute meeting with the patient. The intervention program included education about diagnosis, treatment plan, medications, and the importance of regular medication compliance; a review of follow-up appointments; the provision of pill-boxes and instructions on use; and the provision of handouts about the disease and medications, as well as a list of community resources and support groups according to the diagnosis.
Both groups received follow-up phone calls from a psychiatric nurse 1, 3, 6, and 12 months after the day of discharge to complete the follow-up questionnaires. The pharmacies and doctors of both groups were contacted at the same intervals to obtain an independent measure of the dates the participants filled their medications and whether they kept their follow-up appointments or not.
At 1-year follow-up, information was available for 38 of the original 56 participants. Dropout or loss of contact with the research team was responsible for the 18 withdrawn subjects. The collected data was analyzed using the Statistical Analysis System. Statistical relevance was only assigned to results that had a P value of less than .05.
Overall compliance was defined as a patient reporting taking 100% of all medications, 7 days a week, and keeping all scheduled appointments, as well as the pharmacy and the doctor reporting the patient has been picking up prescriptions and attending all scheduled appointments, respectively. Compliance with medications was defined as the patient taking 100% of the medications, 7 days a week, and the pharmacy confirming that the patient had been picking up their prescriptions. Compliance with scheduled appointments was defined as the patient reporting attending all scheduled appointments with their psychiatrist, and the psychiatrist confirming that.
Our sample consisted of 50% men and 50% women. Seventy-three percent were Caucasian; the remaining were a mix of African-American, Hispanic, and other. Forty-five percent were never married, 30% married, and 23% divorced, separated or widowed. Forty-six percent had private insurance, 20% had Medicaid, and 9% had Medicare.
The patients' reported compliance rates were consistently higher than the numbers reported by pharmacies and physicians. However, this difference was not statistically significant. The compliance rates were fairly consistent throughout the 1 year of follow up. Compliance rates are shown in Tables 3 and 4 (see page 173).
Compliance With Medications: Patient Versus Pharmacy Report
Compliance With Appointments: Patient Versus Physician Report
Several variables were tested for their relationship with compliance. Race was found to be significantly related to compliance. Caucasian patients had a higher probability of taking medications 7 days a week as compared with non-Caucasians (94% versus 79% respectively; P = .01). In addition, Caucasian patients were more likely to be compliant with follow up appointments as compared to non-Caucasians (79% versus 35% respectively; P = .01). Finally, Caucasians were more likely to be compliant overall as compared to non-Caucasians (67% versus 33% respectively; P = .01).
Age was found to be independent of ethnicity and significantly related to compliance. Older patients (older than 40) were more likely to be complaint overall than younger patients (younger than 40) (68% versus 54% respectively; P = .04).
There was no significant relation between sex and any of the compliance measures. However, we found that women were more likely than men to think that they had received greater levels of information about their medicines and condition by their mental health provider. They were also more likely than men to view their relationship with their doctor in a positive light. Sixty one percent of women said that they have received “a lot” of information about their medications, and 64% said that they have received “a lot” of information about their condition, as compared with 39% and 36% of males, respectively (P = .008 and P = .015 respectively). Sixty-six percent of women described the relationship with their doctor as “very good,” compared with only 34% of men (P = .017).
No significant correlations were found between the measurements of compliance and the amount of education received. We had several measures assessing patients attitudes and beliefs. The answers were ordinal in nature (best to worst, none to always, and so on) and could be expressed on a scale from best to worst or lowest to highest. The ratings for each question were summed for each patient and the resultant “belief” score was tested for significant correlation with the three types of compliance. No significant correlations were observed between those who were compliant and those who were not.
Insurance status, marital status, education level, income, psychiatric diagnosis, number of psychiatric medications, history of compliance with medications, and number of nonpsychiatric medications were all not found to be significantly related to compliance.
The most commonly selected subjective reasons for noncompliance with medications were “I stopped because I always forget to take them” and “I stopped because I could not afford them.” For non-compliance with follow-up appointments, the most common chosen answers were “I was too busy to go see the doctor,” “I forgot the appointment date,” “I did not have a ride,” and “financial cost.”
Patients who were compliant had a better perception of their mental health than noncompliant patients (P = .04) (Table 5, see page 174). We did not find any significant difference in the rate of emergency department visits and psychiatric admissions between compliant and noncompliant patients.
Patient Description of Mental Health Versus Compliance
In our study, the intervention was not significantly related to overall compliance. Those who received the intervention were asked about it at 1, 3, 6, and 12 months. An examination of the trending of the answers over time indicates a gradual increase in the “I don't remember” count. Consequently, for these questions, only the data from the first month was used for analysis. Twenty-four percent of patients chose the answer “the intervention was not helpful at all,” 24% chose “it was helpful just a little bit,” 24% chose “it was helpful a lot,” 19% chose “it was very helpful,” and 9% chose “I don't remember receiving any intervention.”
When asked about the most helpful part of the intervention, the majority of patients answered “discussing the illness with them”; when asked about the least helpful part, the majority answered “I don't remember.” When asked about the pillbox, 43% chose the answer “the pillbox was not helpful,” 38% chose “the pillbox was helpful,” and 19% chose “I don't remember.”
For our study, the overall compliance rates during the 1-year course were 67% at 1 month, 74% at 3 months, 48% at 6 months, and 55% at 1 year. These results are similar to those found elsewhere in the literature, where 70% of patients attended the first aftercare appointment following discharge, 61% attended at least 2 weeks of aftercare, and 40% attended 6 months of aftercare.25 Our patients' compliance rates were higher than those reported in the literature that found that following hospital discharge, 35% to 50% of the patients failed to continue treatment.28
Our high compliance rate can be attributed to several factors. First, even though only the experimental group received the intervention on day of discharge, both groups received phone calls at 1, 3, 6 and 12 months after discharge to inquire about their health status and their compliance rate. This focused attention in itself might have been a type of reminder, and thus may have caused the patients to be more compliant. Second, our study included patients who wanted to participate, and these motivated patients tend to be more compliant in general. There were 15 subjects who refused to be in the study, and most of these patients had the diagnosis of personality disorder (mainly borderline). If those patients had been included in the study, they might have decreased our compliance rate.
The high rate of correlation between the compliance rate reported by our patients and the compliance rate as measured by asking pharmacists and doctors suggests that our patient population is reliable. This has an important implication in clinical practice. Calling the pharmacy and the outpatient doctor to inquire about patient compliance can be time-consuming and restricted by privacy regulations such as the Health Insurance Portability and Accountability Act. Knowing that patients are reliable means that simply asking the patients can give an accurate measure of their compliance status.
Our finding that ethnicity was significantly related to compliance is consistent with findings by Kruse and Rohland,26 who found that “white” patients were more likely to keep their post discharge appointment than “nonwhite” patients. However, we found that age was significantly related to compliance after discharge, with older patients being more compliant than younger patients. This finding is inconsistent with the result of a study that found age to not be related to compliance with aftercare appointments following discharge from a psychiatric unit.25
In our study, we did not find any relationship between admission status (voluntary versus involuntary) and overall compliance post discharge. This is similar to earlier results with adult psychiatric inpatients that reported that the perceived coercion of these patients measured during their hospital admission was not related to compliance with medications or appointment attendance following discharge.29
The multifaceted intervention given the day of discharge did not improve compliance with medications and follow-up appointments during the course of 1 year of follow up, which might have several explanations. The small sample size might have affected the power of the study; the intervention itself might have been poorly designed or poorly executed; or the standard of care at our institution might be high enough that no further interventions are needed.
It is hard to compare our intervention to other studies because other studies are fractured into specific interventions, while our study was more comprehensive. In general, a comprehensive review of randomized controlled trials of interventions intended to improve compliance by McDonald et al.30 concluded that “current methods of improving medication adherence for chronic health problems are mostly complex, labor-intensive, and not predictably effective.”30 Looking at psychiatric disorders specifically, the authors found that interventions that included two or more of the following categories were effective in improving compliance: family therapy, compliance counseling, and education. Studies that included only education were unsuccessful, and studies that exclusively focused on family therapy showed mixed results.30
A study by Sharma et al.31 used interventions consisting of having the patients meet with the therapist and attend “outpatient discharge groups” before discharge. They did not find any statistical significance in the rate of attendance of initial post-discharge follow up appointments between the groups that received the interventions and the control groups. However, Youssef32 conducted an experimental study in which a control group was compared to an experimental family-patient teaching program. The experimental group received discussion of the diagnosis, how to recognize relapse signs and what to do if they occur, and community resources available. A significant number of patients from the experimental group attended scheduled appointments following discharge.
Compliance is a complex problem. Our study revealed that some demographic factors were related to compliance, such as ethnicity and age, while others were not. Our multifaceted intervention did not make any difference in the compliance rate with medications and follow-up appointments following hospital discharge; hoiwever, this might be attributable to a design flaw in our study (poorly designed intervention or small sample size) or to high standard of care. Future studies should evaluate our intervention and try to improve on it and test it in a larger sample.
Editor's note: This recurring department, formerly titled “Contemporary Psychiatry,” has been renamed and expanded to cover a wider range of psychiatry-related topics beyond the scope of our monthly single topics. Areas of interest will include forensic psychiatry, patient compliance, managed care, and other subjects of interest to practicing psychiatrists. To suggest a topic or submit an article proposal, please e-mail
- Lowry DA. Issues of non-compliance in mental health. J Adv Nurs. 1998;28(2): 280–287. doi:10.1046/j.1365-2648.1998.00787.x [CrossRef]9725724
- Stoudemire A, Thompson TL 2nd, . Medication noncompliance: systematic approaches to evaluation and intervention. Gen Hosp Psychiatry. 1983;5(4): 233–239. doi:10.1016/0163-8343(83)90001-4 [CrossRef]6662352
- Franson KL, Smith SL. Psychotherapeutic agents in older adults. Compliance: problems and opportunities. Clin Geriatr Med. 1998;14(1):7–16. doi:10.1016/S0749-0690(18)30127-7 [CrossRef]9456332
- Kemppainen JK, Buffum M, Wike G, et al. Psychiatric nursing & medication adherence. J Psychosoc Nurs Ment Health Serv. 2003;41(2):38–49.12613098
- Svarstad BL, Shireman TI, Sweeney JK. Using drug claims data to assess the relationship of medication adherence with hospitalization and costs. Psychiatr Serv. 2001;52(6):805–811. doi:10.1176/appi.ps.52.6.805 [CrossRef]11376229
- Cramer JA, Rosenheck R. Compliance with medication regimens for mental and physical disorders. Psychiatr Serv. 1998;49(2):196–201. doi:10.1176/ps.49.2.196 [CrossRef]9575004
- Carpenter MD, Mulligan JC, Bader IA, et al. Multiple admissions to an urban psychiatric center: a comparative study. Hosp Community Psychiatry. 1985; 36(12):1305–1308.4086005
- Green JH. Frequent rehospitalization and noncompliance with treatment. Hosp Community Psychiatry. 1988; 39(9):963–966.3215645
- Nelson EA, Maruish ME, Axler JL. Effects of discharge planning and compliance with outpatient appointments on readmission rates. Psychiatr Serv. 2000; 51(7):885–889. doi:10.1176/appi.ps.51.7.885 [CrossRef]10875952
- Wilder JF, Plutchnik R, Conte HR. Compliance with psychiatric emergency room referrals. Arch Gen Psychiatry. 1977;34(8):930–933. doi:10.1001/archpsyc.1977.01770200068006 [CrossRef]889416
- Krulee DA, Hales RE. Compliance with psychiatric referrals from a general hospital psychiatry outpatient clinic. Gen Hosp Psychiatry. 1988;10(5):339–45. doi:10.1016/0163-8343(88)90005-9 [CrossRef]3169533
- Jellinek M. Referrals from a psychiatric emergency room: relationship of compliance to demographic and interview variables. Am J Psychiatry. 197;135(2): 209–213.623334
- Matas M, Staley D, Griffin W. A profile of the noncompliant patient: a thirty-month review of outpatient psychiatry referrals. Gen Hosp Psychiatry. 1992; 14(2):124–130. doi:10.1016/0163-8343(92)90037-B [CrossRef]1592248
- Van Dongen CJ. Is the treatment worse than the cure? Attitudes toward medications among persons with severe mental illness. J Psychosoc Nurs Ment Health Serv. 1997;35(3):21–25.9076705
- Ross DJ, Guggenheim FG. Compliance and the health belief model: a challenge for the liaison psychiatrist. Gen Hosp Psychiatry. 1983;5(1):31–35. doi:10.1016/0163-8343(83)90041-5 [CrossRef]6404696
- Benkert O, Graf-Morgenstern M, Hillert A, et al. Public opinion on psychotropic drugs: an analysis of the factor influencing acceptance or rejection. J Nerv Ment Dis. 1997;185(3):151–158. doi:10.1097/00005053-199703000-00004 [CrossRef]9091596
- Rosenberg KP, Bleiberg KL, Koscis J, et al. A survey of sexual side effects among severely mentally ill patients taking psychotropic medications: impact on compliance. J Sex Marital Ther. 2003;29(4): 289–296. doi:10.1080/00926230390195524 [CrossRef]14504017
- Warner LA, Silk K, Yeaton WH, et al. Psychiatrists' and patients' views on drug information sources and medication compliance. Hosp Community Psychiatry. 1994;45(12):1235–1237.7868111
- Britten N. Psychiatry, stigma, and resistance. Psychiatrists need to concentrate on understanding, not simply compliance. BMJ. 1998;317(7164):963–964. doi:10.1136/bmj.317.7164.963 [CrossRef]9765162
- Bostelman S, Callan M, Rolincik LC, et al. A community project to encourage compliance with mental health treatment aftercare. Public Health Rep. 1994; 109(2):153–157.8153265
- Dow MG, Verdi MB, Sacco WP. Training psychiatric patients to discuss medication issues. Effects on patient communication and knowledge of medications. Behav Modif. 1991;15(1):3–21. doi:10.1177/01454455910151001 [CrossRef]2003848
- Kelly GR, Scott JE. Medication compliance and health education among outpatients with chronic mental disorders. Med Care. 1990;28(12):1181–1197. doi:10.1097/00005650-199012000-00006 [CrossRef]2250501
- Waller DA, Altshuler KZ. Perspectives on patient noncompliance. Hosp Community Psychiatry. 1986;37(5):490–492.3699717
- Robinson GL, Gilbertson AD, Litwack L. The effects of a psychiatric patient education to medication program on post-discharge compliance. Psychiatr Q. 1986–87;58(2):113–118. doi:10.1007/BF01064052 [CrossRef]
- Axelrod S, Wetzler S. Factors associated with better compliance with psychiatric aftercare. Hosp Community Psychiatry. 1989;40(4):397–401.2541062
- Kruse GR, Rohland BM. Factors associated with attendance at a first appointment after discharge from a psychiatric hospital. Psychiatr Serv. 2002;53(4): 473–476. doi:10.1176/appi.ps.53.4.473 [CrossRef]11919363
- Bogin DL, Anish SS, Taub HA, Kline GE. The effects of a referral coordinator on compliance with psychiatric discharge plans. Hosp Community Psychiatry. 1984;35(7):702–706.6086487
- Chen A. Noncompliance in community psychiatry: a review of clinical interventions. Hosp Community Psychiatry. 1991;42(3):282–287.1851496
- Rain SD, Williams VF, Robbins PC. Perceived coercion at hospital admission and adherence to mental health treatment after discharge. Psychiatr Serv. 2003;54(1):103–105. doi:10.1176/appi.ps.54.1.103 [CrossRef]12509675
- McDonald HP, Garg AX, Haynes RB. Interventions to enhance patient adherence to medication prescriptions: scientific review. JAMA. 2002;288(22): 2868–2879. doi:10.1001/jama.288.22.2868 [CrossRef]12472329
- Sharma SB, Elkins D, van Sickle A, Roberts CS. Effect of predischarge interventions on aftercare attendance: process and outcome. Health Soc Work. 1995;20(1):15–20. doi:10.1093/hsw/20.1.15 [CrossRef]9154421
- Youssef FA. Discharge planning for psychiatric patients: the effects of a family-patient teaching programme. J Adv Nurs. 1987;12(5):611–616.3320136
Summary of Items on Initial Questionnaire
|Section||Specific Areas Addressed|
|Demographics||Age, sex, ethnicity, insurance status, marital status, level of education, income|
|Clinical information||Psychiatric diagnosis, substance abuse, perceived coercion surrounding hospital admission, number of psychiatric and nonpsychiatric medications discharged on|
|Satisfaction with care||Perceived education given by physician/nurse, unanswered questions about mental health condition, medications, and follow-up appointments, evaluation of relationship with physician and nurses, overall satisfaction with inpatient care|
|Patient's beliefs and attitudes||Perceived change in condition, insight into mental health condition; perceived severity of condition, effectiveness of medications, effect of condition on quality of life, barriers to taking medications, natural course of condition, attitudes toward psychiatrists and psychiatric medications|
|Patient's history with compliance||Past compliance behavior with psychiatric and nonpsychiatric medications.|
Summary of Items on Follow-up Questionnaire
|Section||Specific Areas Addressed|
|Current health status||Perceived mental health, number of admissions to psychiatric hospital and number of visits to emergency department since hospital discharge|
|Compliance||Frequency and portion of prescribed medication actually taken, number of follow-up appointments actually attended, reasons for not taking medications or not attending follow-up appointments if applicable|
|Relationship with physician||Amount of education received about medication and mental health condition, perceived relationship with psychiatrist|
|Pharmacy and physician||Direct contact with pharmacy and physician to inquire about medication refill activity and appointment attendance|
|Evaluation of intervention (if applicable)||Evaluation of intervention, most and least helpful section of intervention, pill box effectiveness|
Compliance With Medications: Patient Versus Pharmacy Report
|Month||Patient Report (%)||Pharmacy Report (%)||Total Patients|
|1||38 (90%)||36 (86%)||42|
|3||37 (97%)||36 (95%)||38|
|6||32 (89%)||31 (86%)||36|
|12||33 (92%)||28 (78%)||36|
Compliance With Appointments: Patient Versus Physician Report
|Month||Patient Report (%)||Physician Report (%)||Total Patients|
|1||34 (81%)||30 (71%)||42|
|3||29 (83%)||29 (83%)||35|
|6||21 (68%)||19 (61%)||31|
|12||27 (79%)||24 (71%)||34|
Patient Description of Mental Health Versus Compliance
|Answer||Total||Overall Compliant N (%)||Overall Noncompliant N (%)||P value*|
|Very poor||1||0 (0)||1 (100)||.04|
| Poor||26||14 (54)||12 (46)|
|Average||34||18 (53)||16 (47)|
| Good||60||40 (67)||20 (33)|
|Very good||13||10 (77)||3 (23)|