OUP user menu

Complaints as indicators of health care shortcomings: which groups of patients are affected?

Susanne Schnitzer, Adelheid Kuhlmey, Holger Adolph, Julie Holzhausen, Liane Schenk
DOI: http://dx.doi.org/10.1093/intqhc/mzs036 476-482 First published online: 9 July 2012

Abstract

Objective Patient complaints about the health care system and medical services are regarded as indicators of shortcomings in health care systems. This article examines the topics of complaint raised most frequently and analyzes which groups of persons were most affected.

Design Quantitative content analysis using a category system. Logistic regression was used for statistical analysis.

Setting and Participants 13 505 letters of complaint directed to the Federal Commissioner for Patient Issues in Germany between 2004 and 2007.

Main outcome measures Letters of complaint covering at least one topic were categorized to a total of 20 topics.

Results The issues most frequently raised were unjust policies (23.8%), refusal or restriction of drugs (23.8%) and refusal or restriction of non-drug treatments (23.9%). The relative proportion of complaints about the physician–patient relationship increased over the period of analysis (over all 4 years: 9.3%). Multivariate analysis showed that complaints about the topics under examination were more likely to be lodged by people with statutory health insurance, people in a precarious financial situation, people with chronic disease or multimorbidity and women.

Conclusions These results provide important insights into shortcomings in the German health care system that should be seen in the context of recent reform measures. Policy makers should be made aware that certain groups of the population are particularly affected by these changes and take steps to ensure that inequalities in the health care system are not exacerbated.

  • complaints
  • Federal Commissioner for Patient Issues
  • social determinants
  • health policies
  • physician–patient relationship

Introduction

Whereas high patient satisfaction is generally assumed to indicate a high-quality health care system, patient complaints about the health care system and medical services are regarded as indicators of health care shortcomings. Some countries therefore use patient complaints to identify areas that warrant attention in the health care system. Like health care systems themselves, complaint systems and complaint management differ across countries. In most cases, patient complaints are addressed to ombudspersons at either the national or the local level, who then make specific recommendations for policy makers [1, 2].

Germany has a number of complaint systems that vary in scope and structure. The health care reform of 2004 led to the establishment of a Federal Commissioner for Patient Issues, who is authorized to address all concerns of patients in Germany. An initial exploratory study analyzed 850 letters addressed to the Commissioner during the first half of 2005 [3]. Although a few international studies have addressed patient complaints in various settings [48], and some studies have explored how complaints influence physician behavior [9, 10], studies specifically describing the groups of patients who lodge complaints or malpractice claims are scarce [1115].

Based on the assumption that patient complaints indicate health care shortcomings, this article aims to identify the topics of complaint raised most frequently in letters directed to the Federal Commissioner for Patient Issues in Germany between 2004 and 2007 and to analyze which groups of patients were most affected. The complaints made are to be seen in the context of reforms that have been implemented in Germany in recent years, as a result of which patients have had to bear an increased portion of the cost of health care [16]. For example, under a law that came into force in early 2004, over-the-counter drugs are no longer covered by statutory health insurance. Moreover, patients have to pay a practice fee of €10 per quarter for every initial contact with a physician that takes place without referral, and it is no longer possible for hardship groups to be fully exempted from co-payment.

Methods

All complaints addressed to the Patient Commissioner in Germany between 2004 and 2007 were analysed in the context of the research project ‘Information for Participatory Health Care’. A detailed description of the methods of data acquisition and coding has been published elsewhere [17]. All complaints were unsolicited letters and e-mails reporting on specific health care problems encountered in practice. All written complaints received between January 2004 and December 2007 (N= 13 505) were subjected to quantitative content analysis using a category system. The unit of analysis was the individual letter. The authors of the letters generally complained about problems they had experienced themselves: In 80.3% of cases, the author and the patient affected were one and the same person. If this was not the case, the sociodemographic characteristics of both the author and the person affected were coded. The following analyses (Tables 13) draw solely on the data of patients. A total of 20 topics of complaint were coded. In the present analyses, we focused on the three most common topics of complaint—unjust policies, refusal or restriction of drugs, and refusal or restriction of non-drug treatments—and their relation to social determinants and health-relevant factors. Because we found the relative proportion of complaints about the physician–patient relationship to have doubled between 2004 and 2007 (from 6.7% in 2004 to 13% in 2007, Table 2), we also included criticism of the physician–patient relationship as a fourth dependent variable. In line with the overall decrease in the number of complaints made over the period of the study, the absolute number of complaints about the physician–patient relationship was much lower in 2007 (n= 272) than in 2004 (n= 480). However, the relative proportion of complaints in this category increased over time, rising to a similar rate as complaints about unjust policies (13.5%) and refusal or restriction of drugs (13.5%) by 2007.

View this table:
Table 1

Letters of complaint (N= 13 505) by year and by social determinants and health-relevant factors of complainants

n%
Letters per year13 505
 2004721553.4
 2005235217.4
 2006183913.6
 2007209915.5
Gender of complainants13 266
 Men637548.1
 Women632747.7
 Couples5644.3
Age of complainants6627
 ≤184917.4
 19–64310046.8
 65–79250037.7
 ≥805368.1
Insurance status11 380
 Statutory insurance10 67693.8
 Private insurance5364.7
 No insurance1381.2
 Other insurance300.3
Region12 805
 Western Germany899470.2
 Eastern Germany254619.9
 Berlin12659.9
Occupational status5979
 Employees123020.6
 Trainees/students3105.2
 Retirees316653.0
 Pensionersa59610.0
 Unemployed/welfare recipients5328.9
 Other1452.4
Financial problems10367.7
Chronic disease466534.5
Multimorbidity12599.3
  • aRecipients of disability pension or reduced earnings capacity pension.

View this table:
Table 2

Individual topics of complaint by social determinants and health-relevant factorsa

Unjust policiesRefusal/restriction of drugsRefusal/restriction of non-drug treatmentsPhysician–patient relationship
n3215321432331256
All23.823.823.99.3
Year (n)3215321432331256
 2004721529.429.922.46.7
 2005235220.420.124.510.8
 2006183918.016.527.813.6
 2007209913.513.525.113.0
Gender (n)2927*305030281225
 Men637522.222.021.08.6
 Women632723.926.026.710.7
Age (n)178616071744603
 ≤1849123.015.943.611.6
 19–64310023.521.126.99.7
 65–79250032.829.522.07.2
 ≥8053622.825.627.211.9
Occupational status (n)174414701383472
 Employees123019.913.623.710.4
 Trainees/students31024.816.532.611.9
 Retirees316633.129.321.16.9
 Pensionersb59631.533.631.96.7
 Unemployed/welfare recipients53235.023.125.09.2
Region (n)3067*309730841194
 Western Germany899423.323.124.89.8
 Eastern Germany254626.128.824.67.2
 Berlin126524.022.818.210.4
Insurance (n)279629632945955
 Private53612.38.015.714.2
 Statutory10 67625.627.426.88.2
Financial problems (n)3215321432331256
 Yes103644.639.729.24.2
 Not stated12 46922.122.523.59.7
Chronic disease (n)3215321432331256
 Yes466528.137.528.46.8
 Not stated884021.516.521.610.6
Multimorbidity (n)3215321432331256
 Yes125929.839.128.611.0
 Not stated12 24623.222.223.59.1
  • All figures in percent unless otherwise indicated (n).

  • aAll results were statistically significant at P< 0.001 unless otherwise indicated.

  • bRecipients of disability pension or reduced earnings capacity pension.

  • *P= 0.02.

View this table:
Table 3

Predicting individual topics of complaint from social determinants and health-relevant variables: odds ratios (and 95% confidence intervals)

Unjust policiesRefusal/restriction of drugsRefusal/restriction of non-drug treatmentsPhysician–patient relationship
Year (Ref = 2007)
 20042.09 (1.60–2.73)2.04 (1.54–2.70)0.60 (0.48–0.76)0.42 (0.30–0.60)
 20051.31 (0.95–1.79)1.25 (0.90–1.75)0.86 (0.66–1.13)1.04 (0.71–1.53)
 20061.32 (0.94–1.86)1.19 (0.83–1.69)0.84 (0.63–1.13)1.13 (0.76–1.69)
Gender (Ref = men)n.s.
 Women1.21 (1.03–1.41)1.25 (1.07–1.45)1.29 (1.01–1.65)
Age (Ref = ≤18)n.s.n.s.
 19–641.04 (0.68–1.59)0.53 (0.38–0.73)
 65–791.54 (1.01–2.33)0.43 (0.31–0.59)
 ≥801.16 (0.72–1.86)0.49 (0.33–0.73)
Occupational status (Ref = employees)n.s.n.s.n.s.
 Trainees/students1.31 (0.76–2.26)
 Retirees1.39 (0.96–2.02)
 Pensioners*1.37 (1.00–1.87)
 Unemployed/welfare recipients1.70 (1.22–2.37)
Region (Ref = Western Germany)n.s.n.s.n.s.
 Eastern Germany1.40 (1.16–1.67)
 Berlin0.99 (0.77–1.27)
Insurance status (Ref = private)
 Statutory2.67 (1.62–4.42)2.84 (1.58–5.08)1.96 (1.26–3.05)0.54 (0.34–0.84)
Financial problems2.27 (1.86–2.76)1.54 (1.26–1.89)1.27 (1.03–1.56)0.45 (0.28–0.72)
Chronic diseasen.s.2.59 (2.21–3.03)1.33 (1.13–1.56)0.72 (0.55–0.96)
Multimorbidityn.s.1.37 (1.12–1.69)1.29 (1.04–1.59)1.86 (1.33–2.61)
  • Ref, reference category; n.s., not significant, variable excluded from the model.

  • N = 3687.

  • aRecipients of disability pension or reduced earnings capacity pension.

It is typical of the source material and the data collection method that not all of the social determinants and health-relevant factors of interest were systematically reported by complainants. Although data on gender (98.2%), region (94.8%) and insurance status (84.3%) were available for a high percentage of complainants, less information was available on age (49.1%) and occupational status (44.3%). However, a structural comparison of ‘responders’ and ‘non-responders’ in a previous analysis showed that the results from the subsample in which these determinants were specified can be generalized to the full sample of complainants [17]. All status variables and health-relevant data that were reported by or could be reconstructed for a statistically sufficient proportion of complainants were used to identify the groups of persons who lodged complaints most often: gender, age, region, occupational status, insurance status, financially precarious situation, chronic disease and multimorbidity. In order to gain insights into the pattern of complaints over time, we further analysed the data by year of complaint (2004–07). Disease was coded as chronic if explicitly reported as such by the complainant or if a disease mentioned in the letter of complaint could be clearly classified as chronic. Complainants who had at least three diseases or explicitly mentioned multimorbidity were coded as having multimorbidity. Those who gave clear indications of being in a difficult financial situation (e.g. attached income statements or reports on their financial situation) were coded as being in a financially precarious situation.

Relations between topics of complaint and social determinants/health-relevant factors are presented in the form of relative proportions; differences in these proportions were statistically analysed using the chi-square test. Backward conditional logistic regression was used to control for confounding variables, and iterative methodology was used to remove nonsignificant independent variables and recalculate the regression model [18]. The results are shown in the tables as odds ratios (ORs) and 95% confidence intervals.

Results

Table 1 presents a breakdown of letters of complaint by sociodemographic determinants and health-relevant factors. The number of letters of complaint per year is also presented. Note that more than half of all complaints made over the study period were lodged in 2004. The health status of the complainants showed a high level of symptomatology: 35% (n= 4665) had a chronic disease and 9% (n= 1259) were identified as having multimorbidity. The high proportion of older people in the sample is also notable: nearly half of the complainants (46%/n= 3036) whose age could be reconstructed were 65 years or older. Accordingly, the percentage of retirees in the sample was disproportionately high (53%/n= 3166).

As mentioned above, the three most common topics of complaint were unjust policies, refusal or restriction of drugs, and refusal or restriction of non-drug treatments. The ‘unjust policies’ category included all complaints relating to aspects of justice; for example, complaints about inequalities in health care provision to the detriment of specific groups (elderly people, chronically ill patients, families and lower-income people) or criticism of ‘two-class medicine.’ Complaints about ‘refusal or restriction of drugs’ were allocated to a second category, and complaints about certain diagnostic methods and the prescription of (non-drug) treatments as well as aids and appliances were subsumed under ‘refusal or restriction of non-drug treatments.’ The fourth topic of analysis—‘criticism of the physician–patient relationship’—subsumed complaints about a lack of diagnostic and therapeutic information; unfriendly, disrespectful treatment; insufficient consultation times; lack of empathy; lack of patient involvement in treatment decisions and lack of confidence in physicians.

The pattern of the topics of complaint varied over time (Table 2). Between 2004 and 2007, the number of complaints about unjust policies and refusal/restriction of drugs decreased, whereas the number of complaints about the physician–patient relationship increased.

Complaints about refusal/restriction of drugs or non-drug treatments were more frequently voiced by people with serious financial problems, patients with chronic disease or multimorbidity and holders of statutory health insurance. For example, only 8% of private policyholders reported problems with the refusal/restriction of drugs, as opposed to 27.4% of statutory policyholders. Furthermore, complaints about the refusal/restriction of both drugs and non-drug treatments were lodged more frequently by women than by men. Likewise, drug prescription was criticized more frequently by eastern than by western Germans. Older people between the ages of 65 and 79 years more frequently made complaints about refusal/restriction of drugs, whereas young people aged 18 or below were more affected by the refusal/restriction of non-drug treatments. These findings are reflected in the results for occupational status: complaints about refusal or restriction of non-drug treatments were most frequently made by trainees/students. Inspection of the subcategories of the ‘refusal/restriction of non-drug treatments’ category revealed that trainees/students most often complained about dental services.

Complaints about unjust policies were also voiced more frequently by people with serious financial problems and by holders of statutory health insurance. Nearly half of those living in a precarious financial situation (44.6%) who lodged a complaint criticized health policy as being unjust. A large proportion of complainants who were unemployed or receiving welfare benefits (35.0%) also criticized the injustice of health policy.

Letters criticizing the physician–patient relationship were most frequently sent by patients with multimorbidity and holders of private health insurance: 14.2% of private policyholders who made a complaint criticized their physician, relative to 8.2% of statutory policyholders. People in a difficult financial situation lodged relatively fewer complaints about the physician–patient relationship than people who did not have or did not report financial problems. Additionally, the physician–patient relationship was criticized less frequently by men than by women and by eastern Germans than by western Germans or residents of Berlin. In terms of age, the oldest age group (80 years and older) and the youngest age group (children and adolescents up to 18 years) voiced most dissatisfaction with the physician–patient relationship.

The bivariate correlations presented in Table 2 were largely confirmed by multivariate logistic regression analysis (Table 3). However, when the other determinants were included in the models, occupational status was no longer significantly associated with complaints about the restriction/refusal of health care services. Moreover, criticism about health policies deemed unjust did not differ between men and women, between eastern and western Germans, across age groups, between patients with and without chronic disease or between patients with and without multimorbidity. Complaints regarding the physician–patient relationship were no longer associated with age, occupational status or region. All other results were in the same direction in the multivariate regression models as in the bivariate analysis. For instance, the probability of complaining about refusal/restriction of drugs was nearly three times higher (OR = 2.84) among people with statutory health insurance than among those with private insurance and more than twice as high (OR = 2.59) among chronically ill people than among those who do not have or did not report a chronic disease. Likewise, criticism about the refusal/restriction of drugs was more likely to be voiced by women, by people in a precarious financial situation, and by patients aged between 65 and 79 years than by the respective reference groups. The probability of complaining about the injustice of health policies was more than twice as high (OR = 2.67) among people with statutory health insurance and among those with financial problems (OR = 2.27) than in the respective reference groups. However, the pattern of results regarding the physician–patient relationship was reversed. The probability of complaining about the physician–patient relationship was lower among people with statutory health insurance than among those with private insurance (OR = 0.54) and among people with financial problems than among those whose financial status was unknown (OR = 0.45).

Discussion

The findings of this study, based on 13 505 letters of complaint addressed to the Federal Commissioner for Patient Issues in Germany between 2004 and 2007, reveal that most complaints related to unjust policies, refusal/restriction of drugs and refusal/restriction of non-drug treatments. Complaints about the physician–patient relationship increased over the period of analysis, and were made mainly by persons with several chronic diseases, by private policyholders and by women. Unjust health policies were primarily criticized by holders of statutory health insurance, by persons with financial problems and by unemployed persons. Likewise, complaints about refusal/restriction of health care services were most often made by holders of statutory health insurance and by persons with financial problems. Patients with one or more chronic diseases and women were also more likely to lodge complaints about health care services than were the respective reference groups. Occupational status played no role here, however. Whereas younger patients were more likely than the other age groups to criticize the refusal/restriction of non-drug treatments, complaints about refusal/restriction of drugs were made significantly more frequently by patients aged between 65 and 79 years. Furthermore, complaints regarding the refusal/restriction of drugs were lodged more frequently by eastern Germans than by western Germans. This finding may be attributable to the lower density of physicians in eastern Germany, which has an unfavorable influence on the overall health care situation in that part of the country.

What other reasons could there be for specific groups criticizing the refusal/restriction of drugs? As the present analysis controlled for chronic disease and multimorbidity, the fact that older persons and women lodged more complaints about drug prescription cannot be attributed to poorer health. Why then did women make more complaints about the refusal/restriction of drugs? Several national and international studies have shown that women are sometimes prescribed different and often prescribed more drugs than men [1921]. Thus, it is possible that women have been more adversely affected by the increasing restriction of drug and non-drug treatments that has resulted from the health care reform measures implemented in Germany in recent years. This possibility requires further research. The finding that patients with chronic disease and multimorbidity complained more about restriction of drug and non-drug treatments is plausible given their higher health care utilization and higher mean drug intake. Patients taking more drugs have been more adversely affected by the implementation of prescription co-payment and by restrictions on the prescription of certain treatments that have been imposed in conjunction with reform measures. With regard to the higher frequency of complaints about the restriction/refusal of drugs among older patients, international studies have shown that the prescription of drugs for the elderly is inadequate [22, 23]. The same holds for Germany, as has been confirmed by a number of regional studies [24, 25].

Complaints about the refusal/restriction of health care services and unjust policies made by patients with financial problems can also be seen in the context of the increasing financial burden that is being placed on patients in the German health care system since the 2004 reforms. People with economic problems regard the reform measures as particularly unjust and are more severely affected by the new charges. This applies not only to low-wage earners and low-income groups but also to unemployed persons and welfare recipients, who were indeed also more likely to criticize unjust health policies.

Apart from persons with financial problems, unemployed persons and welfare recipients, holders of statutory health insurance (86.6% of the German population) [26] have to bear an increased burden of additional payments and coverage restrictions in the wake of the reforms. Accordingly, statutory policyholders made more complaints about health care restrictions and unjust policies than did private policyholders. At the same time, statutory policyholders lodged fewer complaints about the physician–patient relationship than did private policyholders. Numerous previous studies in Germany have found differences between the health care provision for statutory and private policyholders. However, these differences related primarily to services such as waiting times and appointments [27, 28] and not directly to medical treatment. It is possible that private policyholders' more frequent complaints about their physicians are due to higher expectations. In Germany, only a small percentage of the population (12.7%) has private health insurance. Only certain occupation groups (e.g. civil servants, self-employed persons and high earners) are eligible for private health insurance.

More than half of all complaints were lodged in 2004. One possible reason for this pattern is the high media interest in the establishment of the office of Federal Commissioner for Patient Issues during that year. Alternatively, the high volume of complaints in 2004 may be interpreted as a response to the health care reforms that came into effect in early 2004 and the changes they entailed. From 2005, the number of complaints leveled off at a moderate rate of some 2000 letters per year.

As mentioned in the introduction, recourse to the Federal Commissioner for Patient Issues is only one of several complaint systems in Germany. Our study is thus limited in this respect: the analyses reported are representative of all complaints addressed to the Commissioner for Patient Issues, but probably not of all complaints made in Germany. Another limitation is that the groups identified cannot be compared with persons who did not formulate complaints, as the data set contains only complaint cases. Furthermore, it is possible that not all health problems could be reconstructed from the available data—not all complainants reported on their health, and patients were coded as having chronic disease or multimorbidity only if this was explicitly mentioned or clearly indicated. However, it seems likely that people with such health problems would have mentioned them in their letters of complaint.

Despite these limitations, the present results yield important insights into shortcomings in the German health care system that can inform the future development of the country's health policy. Based on complaints data, we were able to identify the topics policyholders perceived to be most urgently in need of attention in the German health care system and the groups of patients most affected. However, it is important to bear in mind that these complaints represent only the tip of the iceberg; for every complaint made, several others probably go uncommunicated. Studies have shown that only a small proportion of those who are unsatisfied with their physician or with the health care system actually lodge a complaint [11, 28]. Problems with serious implications are the subject of more complaints than are problems with less serious implications; likewise, ongoing or recurrent problems are the subject of more complaints than are one-off isolated errors [13]. Accordingly, complaints can largely be regarded as being made patients or groups of patients with a certain level of suffering who feel compelled to make their voices heard.

As described in the introduction, a number of reforms have been implemented in the German health care system in recent years, with far-reaching implications. Policy makers need to take steps to ensure that these reforms do not exacerbate existing inequalities in the health care system. Moreover, they need to mitigate the consequences of the additional financial burden that has been imposed on certain groups. The present analysis offers first insights into the differing effects that the reforms have had on different groups of patients.

Funding

This work was supported by a grant from the Hans Böckler Foundation [S-2007-60-4].

References

View Abstract