Email updates

Keep up to date with the latest news and content from BMC Medical Research Methodology and BioMed Central.

Open Access Research article

Performing meta-analysis with incomplete statistical information in clinical trials

Jianbing Ma1*, Weiru Liu1, Anthony Hunter2 and Weiya Zhang3

Author Affiliations

1 School of Electronics, Electrical Engineering and Computer Science, Queen's University Belfast, Belfast, BT7 1NN, UK

2 Department of Computer Science, University College London, Gower Street, London, WC1E 6BT, UK

3 Academic Rheumatology, Medical & Surgical Sciences, City Hospital, Nottingham, NG5 1PB, UK

For all author emails, please log on.

BMC Medical Research Methodology 2008, 8:56  doi:10.1186/1471-2288-8-56

The electronic version of this article is the complete one and can be found online at: http://www.biomedcentral.com/1471-2288/8/56


Received:12 May 2008
Accepted:18 August 2008
Published:18 August 2008

© 2008 Ma et al; licensee 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.

Abstract

Background

Results from clinical trials are usually summarized in the form of sampling distributions. When full information (mean, SEM) about these distributions is given, performing meta-analysis is straightforward. However, when some of the sampling distributions only have mean values, a challenging issue is to decide how to use such distributions in meta-analysis. Currently, the most common approaches are either ignoring such trials or for each trial with a missing SEM, finding a similar trial and taking its SEM value as the missing SEM. Both approaches have drawbacks. As an alternative, this paper develops and tests two new methods, the first being the prognostic method and the second being the interval method, to estimate any missing SEMs from a set of sampling distributions with full information. A merging method is also proposed to handle clinical trials with partial information to simulate meta-analysis.

Methods

Both of our methods use the assumption that the samples for which the sampling distributions will be merged are randomly selected from the same population. In the prognostic method, we predict the missing SEMs from the given SEMs. In the interval method, we define intervals that we believe will contain the missing SEMs and then we use these intervals in the merging process.

Results

Two sets of clinical trials are used to verify our methods. One family of trials is on comparing different drugs for reduction of low density lipprotein cholesterol (LDL) for Type-2 diabetes, and the other is about the effectiveness of drugs for lowering intraocular pressure (IOP). Both methods are shown to be useful for approximating the conventional meta-analysis including trials with incomplete information. For example, the meta-analysis result of Latanoprost versus Timolol on IOP reduction for six months provided in [1] was 5.05 ± 1.15 (Mean ± SEM) with full information. If the last trial in this study is assumed to be with partial information, the traditional analysis method for dealing with incomplete information that ignores this trial would give 6.49 ± 1.36 while our prognostic method gives 5.02 ± 1.15, and our interval method provides two intervals as Mean ∈ [4.25, 5.63] and SEM ∈ [1.01, 1.24].

Conclusion

Both the prognostic and the interval methods are useful alternatives for dealing with missing data in meta-analysis. We recommend clinicians to use the prognostic method to predict the missing SEMs in order to perform meta-analysis and the interval method for obtaining a more cautious result.

Background

Clinical trials are widely used to test new drugs or to compare the effect of different drugs [2]. For example, many clinical trials have been carried out to investigate the efficacy of drugs for lowering intraocular pressure, such as travoprost, bimatoprost, timolol, and latanoprost, [3-11], and many to investigate the oral diabetes medication for adults with Type-2 diabetes [12-17]. Given that there is a huge number of trials available and reports on trials are very time consuming to read and understand, a systematic review of related trials is useful for medical practioners or other health care professionals to obtain an overall estimation about drugs/therapies of interest. Meta-analysis is the technique commonly used to summarize related trials results. Statistically, meta-analysis is in effect a process for merging sampling distributions (see its explanation in Preliminaries) into a single distribution.

When the full information of sampling distributions used in the trials is available, merging the results from these trials is usually a matter of systematic use of established techniques from statistics. However, in reality, some trial results reported in the literature are statistically incomplete. For instance, the standard error of mean (SEM) may be missing from a sampling distribution. A clinical trial with incomplete information is often abandoned, or, an SEM from a similar trial is used to substitute the missing data. Obviously, abandoning trials will decrease the power of a meta-analysis since trials that are eligible for analysis cannot be taken into account. On the other hand, if these trials are considered and a missing SEM is replaced by a substitute SEM, two major difficulties of this approach are inevitable. First finding a similar trial can be very difficult. Second there is no guarantee that there exists a similar trial with full information. Furthermore, measuring the similarity between trials is another issue that must be addressed. Therefore, an appropriate method is urgently needed to deal with sampling distributions with missing information when considering meta-analysis.

In this paper, we propose a prognostic method where missing SEMs are predicted from known information (SEMs) and an interval method where an interval is calculated in which the missing SEMs are assumed to be. These two methods are based on the assumption that the sampling distributions to be analyzed and merged are from the same population by random selection. In a sense, the two methods can be viewed as extremes. The prognostic method uses an estimate of the missing SEM, while the interval method aims at considering the best and worse cases. Of course, there are other reasonable intermediate procedures, for example, a Bayesian predictive approach is a very natural tool for dealing with missing information.

Significantly, our methods can also be used in the process of merging between group differences (almost all meta-analysis performs merging of between group differences). In fact, our methods can be applied to any results obtained from clinical trials, as long as these results (regarded as random variables) theoretically follow a sampling distribution.

Methods

Study examples

The study examples were drawn from two systematic reviews [1,17]. [17] reported a meta-analysis on drugs for patients with Type-2 diabetes and [1] reported a meta-analysis on lowering intraocular pressure for patients with open angle glaucoma or ocular hypertension.

In order to compare the results obtained from our methods with that from [17], we first attempted to collect all the original information on sampling distributions from each of the clinical trials papers used in [17]. After going through these papers, we found that we could not get the sampling distributions with full information from every single paper, as some of these papers did not provide the SEM values. We then applied our two methods to merge these (including partial) distributions and compared the merged result with that obtained in [17]. In contrast to [17], in [1] every sampling distribution used has full statistical information. To test the adequacy of our methods for dealing with missing information, we randomly selected a trial from [1] and deleted its SEM value. We then applied our two methods to recover this missing SEM value before merging this trial with the rest. Finally, we compared our results with that from [1].

Preliminaries

In statistics, a normal distribution associated with a random variable is denoted as X ~ N (μ, σ2). For the convenience of further calculations in the rest of the paper, we use notation X ~ N (μ, σ) instead of X ~ N (μ, σ2) for a normal distribution of variable X.

In statistics, random samples of individuals are often used as the representatives of the entire group of individuals (often denoted as a population) to estimate the values of some parameters of the population. The mean of variable X of the samples, when the sample size is reasonably large, follows a normal distribution. This distribution is typically referred to as a sampling distribution.

We use X ~ N (μ, SEM) to denote a sampling distribution with mean value μ and standard error of mean SEM.

The basic merging rule for sampling distributions is as follows.

Let X1 ~ N (μ1, SEM1) and X2 ~ N (μ1, SEM2) be two sampling distributions, then the merged sampling distribution X ~ N (μ, SEM) is given by

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M1','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M1">View MathML</a>

(1)

Equation 1 can be easily induced from σ1 = σ2 where σi = SEMi<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M2','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M2">View MathML</a> and ni is the number of samples of the ith trial (for i = 1, 2).

Conventionally, clinicians usually use notation <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M3','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M3">View MathML</a>, thus Equation 1 can be rewritten as:

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M4','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M4">View MathML</a>

(2)

It is easy to see that the merging rule is associative, thus this rule can be straightforwardly used to merge multiple sampling distributions. Let Xi ~ N (μi, SEMi), 1 ≤ i k, then for the merged sampling distribution X ~ N (μ, SEM), we have

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M5','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M5">View MathML</a>

(3)

This is the classical merging method (i.e., meta-analysis) used by clinicians.

The Prognostic Method

We propose a method to predict missing SEMs for sampling distributions from other distributions that have SEM values. Hereafter, we assume that there are k + l trials altogether where k trials are with full information, i.e.,

(μ1, SEM1, n1),..., (μk, SEMk, nk)

and l trials with partial information, i.e.,

(μk+1, nk+1),..., (μk+l, nk+l).

We want to get the merging result of those k + l trials.

If we have both the SEM value and the sample size n from a trial, then we can obtain σ by equation SEM = <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M6','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M6">View MathML</a>. Thus for k trials with

(μ1, SEM1, n1),..., (μk, SEMk, nk)

we get σi = SEMi<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M2','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M2">View MathML</a>, 1 ≤ i k. Because it is assumed that these trials randomly select samples from the same population, these σis suggest what the real value σ could be. In fact, from the well-known Error Theory, we can use

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M7','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M7">View MathML</a>

to replace the real σ. When k gets larger, σ* gets closer to the real σ. Therefore, for a sampling distribution without a standard error, we can use <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M8','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M8">View MathML</a> as its standard error where <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M9','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M9">View MathML</a> is the sample size of this sampling distribution.

To summarize, the prognostic method uses the following equation to predict the missing <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M10','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M10">View MathML</a> value for trial j with sample size <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M9','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M9">View MathML</a>, given that for k trials, each of which has the SEMi value and the sample size ni.

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M11','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M11">View MathML</a>

Then we are able to use the standard meta-analysis method to merge trials with predicted SEM values and trials with full information.

The Interval Method

In contrast to estimating a single value for a missing SEM as discussed above, in this section, we establish a method to estimate a reliable interval for a missing SEM. Then we propose the corresponding merging rule for this case.

For the k trials with

(μ1, SEM1, n1),..., (μk, SEMk, nk),

we have σi = SEMi <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M2','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M2">View MathML</a>, 1 ≤ i k. Let

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M12','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M12">View MathML</a>

It is intuitive to consider that the real σ value is in between σmin and σmax. Therefore for a sampling distribution without the standard error SEM*, we have <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M13','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M13">View MathML</a> where n* is the sample size of this trial.

It remains for us to investigate how to use this interval value for merging. For simplicity and illustration, first let us consider the case where there is only one trial with a missing SEM. Let μk denote the weighted average of μ1 to μk (the k trials with known SEMs), i.e.,

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M14','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M14">View MathML</a>

μk is equivalent to the merged mean value for the k trials in meta-analysis. Let

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M15','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M15">View MathML</a>

and

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M16','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M16">View MathML</a>

then we have the following result.

Let Xi ~ N (μi, SEMi), 1 ≤ i k, denote the ith sampling distribution with sample size ni and Xi+1 ~ N (μi+1, SEMi+1) denote the (k + 1)th sampling distribution with sample size ni+1 where value SEMi+1 is unknown. Then the merged result N (μ, SEM) of the interval method is:

1. If μk+1μk, then μ ∈ [<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M17','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M17">View MathML</a>, <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M18','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M18">View MathML</a>]

2. If μk+1 > μk, then μ ∈ [<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M18','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M18">View MathML</a>, <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M17','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M17">View MathML</a>]

and

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M19','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M19">View MathML</a>

The corresponding bounds of interval for μ depend on whether μk+1 is larger or smaller than μk. If μk+1 is not bigger than μk, then the lower (resp. upper) bound of its interval <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M17','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M17">View MathML</a> (resp. <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M18','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M18">View MathML</a>) is obtained from the result of merging k + 1 trials by the traditional meta-analysis method where the (k + 1)th SEM is assumed to be <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M20','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M20">View MathML</a>. On the other hand, if μk+1 is larger than μk, then the lower (resp. upper) bound is obtained from the result of merging k + 1 trials by the traditional meta-analysis method where the (k + 1)th SEM is assumed to be <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M21','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M21">View MathML</a>

Similarly, for l trials with incomplete statistical information, we define

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M22','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M22">View MathML</a>

and

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M23','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M23">View MathML</a>

i, k + 1 ≤ i k + l, we let

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M24','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M24">View MathML</a>

and

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M25','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M25">View MathML</a>

then we have the following result.

Let Xi ~ N (μi, SEMi), 1 ≤ i k + l, denote the ith sampling distribution with sample size ni such that SEMi is assumed missing when i > k, then the merged result N (μ, SEM) applying the interval method to these k + l trials is:

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M26','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M26">View MathML</a>

and

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M27','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M27">View MathML</a>

Merging of the Differences between Groups

In clinical trials, the between group difference of two drugs/therapies about two groups is frequently used. Suppose that the numbers of patients in the two groups are the same (majority of clinical trials allocate (approximately) the same number of patients in two contrast groups) and we denote this number as n. Assume that the first group gives a sampling distribution of the effect of one drug as X1 ~ N (μ1, SEM1) and the second group about another drug as X2 ~ N (μ2, SEM2). Conventionally, the between group difference (of the two drugs involved) is calculated as

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M28','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M28">View MathML</a>

Thus we get

<a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M29','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M29">View MathML</a>

Let <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M30','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M30">View MathML</a> be a fixed value since σ1 and σ2 are fixed by the selected populations, we get SEM = <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M6','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M6">View MathML</a>. Therefore, the two methods proposed in the previous sections are also suitable in the process of merging the between group differences.

Results and discussion

Case Study of Drugs for Type-2 Diabetes

In this section, we use the data from clinical trials on medication for adults with Type-2 diabetes as our first case study.

Many research papers and reports have been published to show the effectiveness of various oral medication for Type-2 diabetes ([12-16], etc.). For oral medication of Type-2 diabetes, the meta-analysis in [17] compared each pair of drugs on systolic blood pressure (SBP for short), diastolic blood pressure (DBP), low density lipoprotein cholesterol (LDL-C) and high density lipoprotein cholesterol (HDL-C), etc.

In this section, we consider the between group difference on the effectiveness of pairs of drugs for lowering LDL-C. In [17], there were 14 sets of results comparing between group difference of 14 pairs of drugs on lowering LDL-C. We selected four sets among these 14 sets for our case study, because each of them contains more than four trials so that our merging method can be applied more adequately. Our data were obtained from the original papers, most of which are the same as the data used in [17], however some data used in [17] are different from what we have found in the original papers. When this happens, we will state it clearly in the examples.

The drugs being compared in the selected four groups are

1. Thiazolidinedione versus Metformin (Example 1),

2. Triazolidinedione versus second generation Sulfonylureas (Example 2),

3. Metformin versus Metformin plus second generation Sulfonylureas (Example 3),

4. Second generation Sulfonylureas versus Metformin plus second generation Sulfonylureas (Example 4).

Example 1 Drugs for reduction of low density lipoprotein (LDL for short) were studied by many papers. They compared the LDL-C reduction between different trial groups. For example, to compare drugs Thiazolidinedione and Metformin, we obtained the following (in mg/dL) sampling distributions from four trials.

[18]: XPa ~ N (13.3, SEMPa) with n = 102.

[12]: XLT ~ N (7.8, 20.5) with n = 20.

[19]: XHan ~ N (10.1, 2.7) with n = 320.

[20]: XSch ~ N (15.2, SEMSch) with n = 597.

Here n is the size of samples (number of patients) in each group of the trial, and SEMPa and SEMSch stand for the missing values (SEM values) from their respective trials data.

The meta-analysis results obtained by using prognostic method, interval method, and traditional method for incomplete trials information are provided in Table 1.

Table 1. Thiazolidinedione vs. Metformin

In Table 1, the first column shows the SEMs of XPa and XSch are missing. The second column shows that the result obtained by the prognostic (P for short) method is Xp ~ N (12.56, 1.91) (denoted as 12.56 ± 1.91 in Table 1). The third column stands for the result obtained by the interval method (Int for short) and the last column stands for the result given by the traditional method on meta-analysis with incomplete trials information (Inc for short), i.e., removing trials with incomplete information. The rest of the tables in this case study follow the same notations and explanations.

Compared to XBW ~ N (12.5,1.89) [17]which is the meta-analysis result when all information is complete, obviously, with two SEM values missing, XP is still very close to XBW . Therefore, XP can be used to replace XBW for this set of trials. In addition, we also have 12.5 ∈ [11.94, 13.39] and 1.89 ∈ [1.54, 2.14] which indicate that XI is a good estimation.

Example 2 To compare Thiazolidinedione and second generation Sulfonylureas, we obtained the following sampling distributions (in mg/dL) from five trials.

[12]: XLT ~ N (10.5, 14.44) with n = 20.

[13]: XCM ~ N (11.31, 1.59) with n = 620.

[14]: XTJ ~ N (6.63, SEMTJ) with n = 100.

[15]: XPM ~ N (5, 6.04) with n = 86.

[16]: XMC ~ N (14.6, SEMMC) with n = 315.

There are two missing SEM values.

The meta-analysis results for this example are provided in Table 2.

Table 2. Triazolidinedione vs. second generation Sulfonylureas

In this example, XP is reasonably close to XBW where XBW ~ N (10.4,1.61) [17]. However XInc is closer.

Example 3 To compare Metformin and Metformin plus second generation Sulfonylureas, we looked at another set containing six trials with sampling distributions as follows (in mg/dL).

[21]: XGa ~ N (-10.2, SEMGa) with n = 168.

[22]: XGo ~ N (-7.0, 5.18) with n = 60.

[23]: XMa ~ N (-3.9, 4.11) with n = 103.

[24]: XDeF ~ N (2.0, 2.8) with n = 186.

[25]: XH94 ~ N (-3.1, 3.6) with n = 30.

[26]: XH91 ~ N (8.2, 5.8) with n = 15.

There is only one missing SEM value. The meta-analysis result in [17] is XBW ~ N (-1.6, 2.53) while we obtained the following results.

The meta-analysis results for this example are provided in Table 3.

Table 3. Metformin vs. Metformin plus second generation Sulfonylureas

Some data (sampling distributions of [23,25,26]) used in [17] are somehow different from the data we found from the original papers. Naturally, there is a margin between the merged sampling distribution from our methods and that from the meta-analysis. Therefore, it is difficult to compare these results.

Example 4 A set of seven trials was analyzed in [17] to compare the second generation Sulfonylureas versus Metformin plus second generation Sulfonylureas. We obtained the following seven sampling distributions (in mg/dL) from them.

[21]: XGa ~ N (-2.2, SEMGa) with n = 168.

[22]: XGo ~ N (-0.2, 4.6) with n = 60.

[23]: XMa ~ N (3.9, 4.72) with n = 103.

[27]: XGr ~ N (10.14, 10.17) with n = 87.

[24]: XDeF ~ N (11.0, 2.83) with n = 186.

[25]: XH94 ~ N (7.41, 4.22) with n = 30.

[26]: XH91 ~ N (19.89, 5.54) with n = 15.

There is one SEM value missing. The results from our methods are given in Table 4.

Table 4. Second generation Sulfonylureas vs. Metformin plus second generation Sulfonylureas

It appears that XP is not so close to XBW where XBW ~ N (8.1,2.55) [17]. Once again, some data (sampling distributions of [23,25,26]) used in [17] are not the same as given in the original papers. So effectively, we had to use different data to perform the merging of these several trials.

Case Study of IOP Reduction

In the above subsection, we applied our methods to approximate the traditional meta-analysis when some SEM values are missing. Because some of the data we found in the original papers are different from that used in the meta-analysis paper [17], it is hard to judge our approaches in some cases. In order to further validate our approaches experimentally, in this section, we consider and apply our methods to a set of trials with full statistical information. In other words, every trial we consider will have both the mean and the SEM value in its sampling distribution. To validate how accurate our two approaches are for predicting an SEM value, we deleted an SEM value from a trial selected randomly from a set of trials, and applied our methods to predict it. We then applied the meta-analysis method to merge the trial with the predicted SEM value together with the rest of trials in the group to see how close this new result is to the original meta-analysis result. Furthermore, as the traditional method for trials with incomplete information always abandons trials with incomplete information, we also compared our methods with this traditional method.

We considered four sets of trials used in [1] on IOP reduction. [1] reported the meta-analysis of some trials about between group difference of IOP reduction from baselines using drugs Latanoprost and Timolol.

Example 5 (one week results) Three papers on clinical trials about IOP reduction gave the following three sampling distributions comparing Latanoprost and Timolol for one-week trial duration.

[28]: X1 ~ N (8.5, 8.24) with n = 46.

[29]: X2 ~ N (5.62, 7.91) with n = 15.

[30]: X3 ~ N (6.85, 4.01) with n = 20.

The traditional meta-analysis in [1] produced X1w ~ N (6.90, 3.28). We deleted the SEM value from one of these trials in turn and applied our methods to predict it. Then we calculated the merged sampling distribution involving this predicted SEM value. The results are summarized in Table 5.

Table 5. Latanoprost vs. Timolol: one week

In Table 5, the second row starting with "X1" means that if the SEM value of X1 is missing, then from the prognostic method, we get <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M34','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M34">View MathML</a> ~ N (7.55, 2.53), from the interval method, we get <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M35','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M35">View MathML</a> ~ N ([7.33, 7.83], [2.12, 2.80]), and from the traditional method for trials with incomplete information, we get <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M36','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M36">View MathML</a> ~ N (6.60, 3.57). Other rows are explained similarly. We can see that <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M31','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M31">View MathML</a>s, 1 ≤ i ≤ 3, are close to the result obtained by the traditional method.

Example 6 (one month results) Three papers on clinical trials about IOP reduction gave the following three sampling distributions comparing Latanoprost and Timolol for one month duration.

[28]: X1 ~ N (11.0, 6.05) with n = 46.

[31]: X2 ~ N (5.2, 1.66) with n = 184.

[32]: X3 ~ N (0.4, 2.17) with n = 294.

The traditional meta-analysis gives X1m ~ N (3.77, 1.29). The results are summarized in Table 6. <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M31','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M31">View MathML</a>s, 1 ≤ i ≤ 3, are also very close to the result obtained by the traditional method.

Table 6. Latanoprost vs. Timolol: one month

Example 7 (three months results) Five papers on clinical trials about IOP reduction gave the following five sampling distributions comparing Latanoprost and Timolol for three months duration.

[33]: X1 ~ N (4.0, 3.11) with n = 267.

[34]: X2 ~ N (5.41, 5.56) with n = 60.

[35]: X3 ~ N (3.2, 4.03) with n = 36.

[31]: X4 ~ N (7.8, 3.98) with n = 184.

[32]: X5 ~ N (1.9, 2.12) with n = 294.

The traditional meta-analysis gives X3m ~ N (3.52, 1.44). The results are summarized in Table 7. Once again, <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M31','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M31">View MathML</a>s, 1 ≤ i ≤ 5, are close to the result obtained by the traditional method.

Table 7. Latanoprost vs. Timolol: three months

Example 8 (six months results) Four papers on clinical trials about IOP reduction gave the following four sampling distributions comparing Latanoprost and Timolol for six months duration.

[33]: X1 ~ N (4.8, 3.14) with n = 267.

[36]: X2 ~ N (7.1, 1.58) with n = 268.

[35]: X3 ~ N (4.5, 5.23) with n = 36.

[32]: X4 ~ N (1.5, 2.14) with n = 294.

The traditional meta analysis gives X3m ~ N (5.05, 1.15). The results are summarized in Table 8. Finally, <a onClick="popup('http://www.biomedcentral.com/1471-2288/8/56/mathml/M31','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2288/8/56/mathml/M31">View MathML</a>s, 1 ≤ i ≤ 4, are also close to the result obtained by the traditional method.

Table 8. Latanoprost vs. Timolol: six months

By comparing these results, we can conclude that the more trials we have in an example, the closer our result is to the traditional meta-analysis method. This is because when the number of trials included in a meta-analysis gets larger, the error of meta-analysis gets smaller. Therefore, the predicted SEM value from the prognostic method gets closer to the real SEM value and the intervals used in the interval method have a better chance to contain the real values.

Discussion

Here we mainly analyze the results obtained from the IOP examples. Since we have stated that some data used in [17] are different from the data we found in the original papers, it would be difficult to compare those results objectively.

First, we found that the results obtained from the prognostic method are closer to the true results. It implies that the prognostic method is truly applicable. Although the interval method gives an interval instead of a single value, the interval provides a clear indication as to where the real value could be, and so it is an alternative to supplement the missing value.

Second, we found that the prognostic method is superior to the traditional method for trials with incomplete information which abandons trials with missing information. In fact, we can see that although in some cases, the results obtained by the traditional method for trials with incomplete information are also close to the true results, in some other cases, they deviate from the true results significantly, such as cases like in Table 6, when the SEM of X2 is missing, in Table 7, that of X4 or X5 is missing, and in Table 8, that of X2 or X4 is missing. That is, the prognostic method is more stable and precise than the traditional method for incomplete trial information. This result is not surprising because the prognostic method takes into account all the eligible trials. On the contrary, if the SEM value of a trial with a large/small mean value is missing, then the traditional method for trials with incomplete information will abandon this trial and naturally gets a result with smaller/higher mean value than the true one. However, the prognostic method does not abandon this trial, and instead it predicts a value for the missing one. Therefore, the final combined result is closer to the real result of meta-analysis. This is also illustrated by Examples 5,6,7,8, etc, except the case of the missing SEM of X1 in Example 5.

Dealing with missing data in statistics, especially in meta-analysis is a very important issue (e.g., [37-39]). However, there are hardly any papers focusing on missing standard errors. This paper thus provides some fresh results about how to deal with this situation. Our assumption that the populations should be the same or similar intuitively follows the well known Missing-at-random (MAR) assumption, except that MAR mainly focuses on repeatedly measured data while ours are focusing on different trials. With this assumption, the prognostic and the interval methods are easily induced. Generally speaking, the prognostic method is related to the regression imputation and the interval method is to some extent a concrete implementation of the best/worst case analysis [40] with our assumption and situations.

A small caveat of the prognostic and the interval methods is that when the input data are imprecise (as shown in Examples 3 and 4), we may obtain worse results than that from the traditional method for incomplete information. This is because when some input data are imprecise, the prognostic method will get wrong predictions for missing SEMs, hence obtain worse meta-analysis results whilst the traditional method for incomplete trial information simply abandons trials with incomplete information, avoiding any wrong predictions and therefore may obtain better results. In addition, in Example 5, when the SEM value of X1 is missing, we found that the result obtained by the prognostic method is not as accurate as that obtained by the traditional method for incomplete information. This is because the SEM value of X1 is exceptionally larger than the others, i.e., as the sample size of X1 is greater than those of X2 and X3, it is expected (according to our assumption of same population) that the SEM of X1 should be smaller, however, it is not. So the prognostic method performs less effectively here.

Conclusion

Methods to deal with statistical information in clinical trials for which the SEMs values are missing is a neglected area in the literature. An obvious solution is to ignore a trial with missing SEM values. Our methods provide an application tool to solve this problem by effectively predicting missing SEMs or providing applicable intervals for the missing SEMs. The prognostic method can be used to obtain a result in a commonly accepted format, e.g., sampling distribution, whilst the interval method, although seemingly not so straightforward, provides a safer and also informative result.

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

JM, WL and AH contributed to the analysis methods and the collection of materials on case studies. WZ contributed to the analysis methods and the evaluation of these new methods with respect to the traditional meta-analysis method and the traditional method for dealing with missing values. All authors read and approved the final manuscript.

Acknowledgements

We would like to thank the UK EPSRC for the project grant supports to Jianbing Ma, Weiru Liu and Anthony Hunter, and Joanna Rawmowski for her logistic support for this project.

References

  1. Zhang WY, Po AL, Dua HS, Azuara-Blanco A: Meta-analysis of randomised controlled trials comparing latanoprost with timolol in the treatment of patients with open angle glaucoma or ocular hypertension.

    Br J Ophthalmol 2001, 85:983-990. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  2. Greenhalgh T: How to Read a Paper: The Basics of Evidence-Based Medicine. BMJ Press; 1997. OpenURL

  3. Chiselita D, Antohi I, Medvichi R, Danielescu C: Comparative analysis of the efficacy and safety of latanoprost, travoprost and the fixed combination timolol-dorzolamide; a prospective, randomized, masked, cross-over design study.

    Oftalmologia 2005, 49(3):39-45. PubMed Abstract OpenURL

  4. Cantor LB, Hoop J, Morgan L, Wudunn D, Catoira Y: Bimatoprost-Travoprost Study Group, Intraocular pressure-lowering efficacy of bimatoprost 0.03% and travoprost 0.004$ in patients with glaucoma or ocular hypertension.

    Br J Ophthalmol 2006, 90(11):1370-1373. PubMed Abstract | Publisher Full Text OpenURL

  5. Gracia-Feijo J, Martinez-de-la Casa JM, Castillo A, Mendez C, Fernandez-Vidal A, Garcia-Sanchez J: Circadian IOP-lowering efficacy of travoprost 0.004$ ophthalmic solution compared to latanoprost 0.005%.

    Curr Med Res Opin 2006, 22(9):1689-1697. PubMed Abstract | Publisher Full Text OpenURL

  6. Howard S, Silvia ON, Brian E, John S, Sushanta M, Theresa A, V M: The Safety and Efficacy of Travoprost 0.004%/Timolol 0.5% Fixed Combination Ophthalmic Solution.

    Ame J Ophthalmology 2005, 140(1):1-8. OpenURL

  7. Michael T, David W, Alan L: Projected impact of travoprost versus timolol and latanoprost on visual field deficit progression and costs among black glaucoma subjects.

    Trans Am Ophthalmol Soc 2002, 100:109-118. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  8. Noecker RJ, Earl ML, Mundorf TK, Silvestein SM, P PM: Comparing bimatoprost and travoprost in black Americans.

    Curr Med Res Opin 2006, 22(11):2175-2180. PubMed Abstract | Publisher Full Text OpenURL

  9. Nicola C, Michele V, Tiziana T, Francesco C, Carlo S: Effects of Travoprost Eye Drops on Intraocular Pressure and Pulsatile Ocular Blood Flow: A 180-Day, Randomized, Double-Masked Comparison with Latanoprost Eye Drops in Patients with Open-Angle Glaucoma.

    Curr Ther Res 2003, 64(7):389-400. Publisher Full Text OpenURL

  10. Parmarksiz S, Yuksel N, Karabas VL, Ozkan B, Demirci G, Caglar Y: A comparison of travoprost, la-tanoprost and the fixed combination of dorzolamide and timolol in patients with pseudoexfoliation glaucoma.

    Eur J Ophthalmol 2006, 16(1):73-80. PubMed Abstract OpenURL

  11. Stefan C, Nenciu A, Malcea C, Tebeanu E: Axial length of the ocular globe and hypotensive effect in glaucoma therapy with prostaglandin analogs.

    Oftalmologia 2005, 49(4):47-50. PubMed Abstract OpenURL

  12. Lawrence J, Reid J, Taylor G, Stirling C, J R: Favorable Effects of Pioglitazone and Metformin Compared With Gliclazide on Lipoprotein Subfractions in Overweight Patients With Early Type 2 Diabetes.

    Diabetes Care 2004, 27(1):41-46. PubMed Abstract | Publisher Full Text OpenURL

  13. Charbonnel BH, Matthews DR, Schernthaner G, Hanefeld M, Brunetti P, the QUARTET Study Group: A long-term comparison of pioglitazone and gliclazide in patients with Type 2 diabetes mellitus: a randomized, double-blind, parallel-group comparison trial.

    Diabetic Medicine 2004, 22:399-405. Publisher Full Text OpenURL

  14. Tan MH, Johns D, Strand J, Halse J, Madsbad S, Eriksson JW, Clausen J, Konkoy CS, Herz M, the GLAC Study Group: Sustained effects of pioglitazone vs. glibenclamide on insulin sensitivity, glycaemic control, and lipid profiles in patients with Type 2 diabetes.

    Diabetic Medicine 2004, 21:859-866. PubMed Abstract | Publisher Full Text OpenURL

  15. Pfüzner A, Marx N, Lüben G, Langenfeld M, Walcher D, Konrad T, Forst T: Improvement of Cardiovascular Risk Markers by Pioglitazone Is Independent From Glycemic Control Results From the Pioneer Study.

    Journal of the American College of Cardiology 2005, 45(12):1925-1931. PubMed Abstract | Publisher Full Text OpenURL

  16. Matthews DR, Charbonnel BH, Hanefeld M, Brunetti P, G S: Long-term therapy with addition of piogli-tazone to metformin compared with the addition of gliclazide to metformin in patients with type 2 diabetes: a randomized, comparative study.

    Diabetes Metab Res Rev 2005, 21:167-174. PubMed Abstract | Publisher Full Text OpenURL

  17. Bolen S, Wilson L, Vassy J, Feldman L, Yeh J, Marinopoulos S, Wilson R, Cheng D, Wiley C, Selvin E, Malaka D, Akpala C, Brancati F, Bass E: Comparative effectiveness and safety of oral diabetes medications for adults with type 2 diabetes.

    Ann Intern Med 2007, 147(6):386-399. PubMed Abstract OpenURL

  18. Pavo I, Jermendy G, Varkonyi TT, Kerenyi Z, Gyimesi A, Shoustov S, Shestakova M, Herz M, Johns D, Schluchter BJ, Festa A, Tan MH: Effect of Pioglitazone Compared with Metformin on Glycemic Control and Indicators of Insulin Sensitivity in Recently Diagnosed Patients with Type 2 Diabetes.

    The Journal of Clinical Endocrinology & Metabolism 2003, 88(4):1637-1645. Publisher Full Text OpenURL

  19. Hanefeld M, Brunetti P, Schernthaner GH, Matthews DR, Charbonnel B: HobotQSG: One-Year Glycemic Control With a Sulfonylurea Plus Pioglitazone Versus a Sulfonylurea Plus Metformin in Patients With Type 2 Diabetes.

    Diabetes Care 2004, 27(1):141-147. PubMed Abstract | Publisher Full Text OpenURL

  20. Schernthaner G, Matthews DR, Charbonnel B, Hanefeld M: Brunetti PoBotQSG: Efficacy and Safety of Pioglitazone Versus Metformin in Patients with Type 2 Diabetes Mellitus: A Double-Blind, Randomized Trial.

    The Journal of Clinical Endocrinology & Metabolism 2004, 89(12):6068-6076. Publisher Full Text OpenURL

  21. Garber AJ, Donovan DS, Dandona P, Bruce S, Park JS: Efficacy of Glyburide/Metformin Tablets Compared with Initial Monotherapy in Type 2 Diabetes.

    The Journal of Clinical Endocrinology & Metabolism 2003, 88(8):3598-3604. Publisher Full Text OpenURL

  22. Goldstein BJ, Pans M, Rubin CJ: Multicenter, Randomized, Double-Masked, Parallel-Group Assessment of Simultaneous Glipizide/Metformin as Second-Line Pharmacologic Treatment for Patients with Type 2 Diabetes MellitusThat Is Inadequately Controlled by a Sulfonylurea.

    Clinical Therapeutics 2003, 25(3):890-903. PubMed Abstract | Publisher Full Text OpenURL

  23. Marre M, Howlett H, Lehertt P, Allavoine T: Improved glycaemic control with metformin glibenclamide combined tablet therapy (Glucovance) in Type 2 diabetic patients inadequately controlled on metformin.

    Diabetic Medicine 2002, 19:673-680. PubMed Abstract | Publisher Full Text OpenURL

  24. Defronzo RA: Goodman AMfTMMSG: Efficacy Of Metformin In Patients With Non-Insulin-Dependent Diabetes Mellitus.

    The New England Journal Of Medicine 1995, 333(9):541-549. PubMed Abstract | Publisher Full Text OpenURL

  25. Hermann LS, Schersten B, Bitzen PO, Kjellstrom T, Lindgarde F, Melander A: Therapeutic Comparision of Metformin and Sulfonylurea, alone and in Various Combinations. A Double-blind Controlled Study.

    Diabetes Care 1994, 17(10):1100-1109. PubMed Abstract | Publisher Full Text OpenURL

  26. Hermann LS, Kjellstrom T, Nilsson-Ehle P: Effects of Metformin and Glibenclamide alone and in Combination on Serum Lipids and Lipoproteins in Patients with Non-insulin-dependent Diabetes Mel-litus.

    Diabete Metab 1991, 17(1 pt 2):174-179. PubMed Abstract OpenURL

  27. Gregorio F, Ambrosi F, Manfrini S, Velussi M, Carle F, Testa R, Merante D, Filipponi P: Poorly Controlled Elderly Type 2 Diabetic Patients: the Effects of Increasing Sulphonylurea Dosages or Adding Metformin.

    Diabet Med 1999, 16(12):1016-1024. PubMed Abstract | Publisher Full Text OpenURL

  28. Diestelhorst M, Almegard B: Comparison of two fixed combinations of latanoprost and timolol in open-angle glaucoma.

    Graefes Arch Clin Exp Ophthalmol 1998, 236:577-581. PubMed Abstract | Publisher Full Text OpenURL

  29. Nicolela MT, Buckley AR, Walman BE, Drance SM: A comparative study of the effects of timolol and latanoprost on blood flow velocity of the retrobulbar vessles.

    Am J Ophthalmol 1996, 122:784-789. PubMed Abstract OpenURL

  30. Rulo AH, Greve EL, Hoyng PF: Additive effect of latanoprost, a prostaglandin F2α analogue, and timolol in patients with elevated intraocular pressure.

    Br J Ophthalmol 1994, 78:899-902. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  31. Mishima HK, Masuda K, Kitazawa Y, Azuma I, Araie M: A comparison of latanoprost and timolol in primary open-angle glaucoma and ocular hypertension: a 12-week study.

    Arch Ophthalmol 1996, 114:929-932. PubMed Abstract OpenURL

  32. Watson P, Stjernschantz J: A six-month, randomized, double-masked study comparing latanoprost with timolol in open-angle glaucoma and ocular hypertension.

    Ophthalmology 1996, 103:126-137. PubMed Abstract OpenURL

  33. Alm A, Stjernschantz J, the Scandinavian Latanoprost Study Group: Effects on intraocular pressure and side effects of 0.005% latanoprost applied once daily, evening or morning-a comparison with tim-olol.

    Ophthalmology 1995, 102:1743-1752. PubMed Abstract OpenURL

  34. Aquino MV, Lat-Luna M: The effect of latanoprost vs timolol on intraocular pressure in patients with glaucoma and ocular hypertension.

    Asian J Ophthalmology 1999, 1:3-7. OpenURL

  35. Mastropasqua L, Carpineto P, Ciancaglini M, Gallenga PE: A 12-month, randomized, double-masked study comparing latanoprost with timolol in pigmentary glaucoma.

    Ophthalmology 1999, 106:550-555. PubMed Abstract | Publisher Full Text OpenURL

  36. Camras CB, the US Latanoprost Study Group: Comparison of latanoprost and timolol in patients with ocular hypertension and glaucoma. A six-month, masked, multicenter trial in the United States.

    Ophthalmology 1996, 103:138-147. PubMed Abstract OpenURL

  37. Lu G, Copas JB: Missing at Random, Likelihood Ignorability and Model Completeness.

    The Annals of Statistics 2004, 32(2):754-765. Publisher Full Text OpenURL

  38. Copas JB, Eguchi S: Local model uncertainty and incomplete-data bias.

    J R Statist Soc B 2005, 67(4):459-513. Publisher Full Text OpenURL

  39. White I: Missing data and departures from randomised treatment in pragmatic trials. [http://www.mrc-bsu.cam.ac.uk/BSUsite/Research/Section11.shtml] webcite

  40. Gamble C, Hollis S: Uncertainty method improved on best-worst case analysis in a binary meta-analysis.

    Journal of Clinical Epidemiology 2005, 58(6):579-588. PubMed Abstract | Publisher Full Text OpenURL

Pre-publication history

The pre-publication history for this paper can be accessed here:

http://www.biomedcentral.com/1471-2288/8/56/prepub